feat(i18n): add internationalization support for runtime

- Implemented translation functionality in help.py, setunset.py, and sid.py using a new i18n module.
- Created locale files for English (en-US) and Simplified Chinese (zh-CN) with appropriate translations.
- Updated default configuration to include language settings.
- Enhanced content safety check strategies to support localized error messages.
- Added language selection feature in the dashboard with prompts for language change confirmation and restart.

.github/PULL_REQUEST_TEMPLATE.md

@@ -3,8 +3,8 @@
 
 ### Modifications / 改动点
 
-<!--请总结你的改动：哪些核心文件被修改了？实现了什么功能？-->
 <!--Please summarize your changes: What core files were modified? What functionality was implemented?-->
+<!--请总结你的改动：哪些核心文件被修改了？实现了什么功能？-->
 
 - [x] This is NOT a breaking change. / 这不是一个破坏性变更。
 <!-- If your changes is a breaking change, please uncheck the checkbox above -->
@@ -21,23 +21,14 @@
 <!--If merged, your code will serve tens of thousands of users! Please double-check the following items before submitting.-->
 <!--如果分支被合并，您的代码将服务于数万名用户！在提交前，请核查一下几点内容。-->
 
-- [ ] 😊 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。
-  / If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc.
+- [ ] 😊 If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc. 
+  / 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。
 
-- [ ] 👀 我的更改经过了良好的测试，**并已在上方提供了“验证步骤”和“运行截图”**。
-  / My changes have been well-tested, **and "Verification Steps" and "Screenshots" have been provided above**.
+- [ ] 👀 My changes have been well-tested, **and "Verification Steps" and "Screenshots" have been provided above**.
+  / 我的更改经过了良好的测试，**并已在上方提供了“验证步骤”和“运行截图”**。
 
-- [ ] 🤓 我确保没有引入新依赖库，或者引入了新依赖库的同时将其添加到 `requirements.txt` 和 `pyproject.toml` 文件相应位置。
-  / I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in `requirements.txt` and `pyproject.toml`.
+- [ ] 🤓 I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in `requirements.txt` and `pyproject.toml`.
+  / 我确保没有引入新依赖库，或者引入了新依赖库的同时将其添加到 `requirements.txt` 和 `pyproject.toml` 文件相应位置。
 
-- [ ] 😮 我的更改没有引入恶意代码。
-  / My changes do not introduce malicious code.
-
-- [ ] ⚠️ 我已认真阅读并理解以上所有内容，确保本次提交符合规范。
-  / I have read and understood all the above and confirm this PR follows the rules.
-
-- [ ] 🚀 我确保本次开发**基于 dev 分支**，并将代码合并至**开发分支**（除非极其紧急，才允许合并到主分支）。
-   / I confirm that this development is **based on the dev branch** and will be merged into the **development branch**, unless it is extremely urgent to merge into the main branch.
-
-- [ ] ⚠️ 我**没有**认真阅读以上内容，直接提交。
-  / I **did not** read the above carefully before submitting.
+- [ ] 😮 My changes do not introduce malicious code.
+  / 我的更改没有引入恶意代码。

.github/workflows/coverage_test.yml

@@ -40,6 +40,7 @@ jobs:
           pytest --cov=astrbot -v -o log_cli=true -o log_level=DEBUG
 
       - name: Upload results to Codecov
-        uses: codecov/codecov-action@v5
+        if: github.repository == 'AstrBotDevs/AstrBot'
+        uses: codecov/codecov-action@v6
         with:
           token: ${{ secrets.CODECOV_TOKEN }}

.github/workflows/dashboard_ci.yml

@@ -8,6 +8,7 @@ on:
 
 jobs:
   build:
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
@@ -45,7 +46,7 @@ jobs:
 
       - name: Create GitHub Release
         if: github.event_name == 'push'
-        uses: ncipollo/release-action@v1.20.0
+        uses: ncipollo/release-action@v1.21.0
         with:
           tag: release-${{ github.sha }}
           owner: AstrBotDevs

.github/workflows/docker-image.yml

@@ -11,7 +11,7 @@ on:
 
 jobs:
   build-nightly-image:
-    if: github.event_name == 'schedule'
+    if: github.repository == 'AstrBotDevs/AstrBot' && github.event_name == 'schedule'
     runs-on: ubuntu-latest
     env:
       DOCKER_HUB_USERNAME: ${{ secrets.DOCKER_HUB_USERNAME }}
@@ -70,14 +70,14 @@ jobs:
         uses: docker/setup-buildx-action@v4.0.0
 
       - name: Log in to DockerHub
-        uses: docker/login-action@v4.0.0
+        uses: docker/login-action@v4.1.0
         with:
           username: ${{ secrets.DOCKER_HUB_USERNAME }}
           password: ${{ secrets.DOCKER_HUB_PASSWORD }}
 
       - name: Login to GitHub Container Registry
         if: env.HAS_GHCR_TOKEN == 'true'
-        uses: docker/login-action@v4.0.0
+        uses: docker/login-action@v4.1.0
         with:
           registry: ghcr.io
           username: ${{ env.GHCR_OWNER }}
@@ -109,7 +109,7 @@ jobs:
         run: echo "Test Docker image has been built and pushed successfully"
 
   build-release-image:
-    if: github.event_name == 'workflow_dispatch' || (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v'))
+    if: github.repository == 'AstrBotDevs/AstrBot' && (github.event_name == 'workflow_dispatch' || (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v')))
     runs-on: ubuntu-latest
     env:
       DOCKER_HUB_USERNAME: ${{ secrets.DOCKER_HUB_USERNAME }}
@@ -169,14 +169,14 @@ jobs:
         uses: docker/setup-buildx-action@v4.0.0
 
       - name: Log in to DockerHub
-        uses: docker/login-action@v4.0.0
+        uses: docker/login-action@v4.1.0
         with:
           username: ${{ secrets.DOCKER_HUB_USERNAME }}
           password: ${{ secrets.DOCKER_HUB_PASSWORD }}
 
       - name: Login to GitHub Container Registry
         if: env.HAS_GHCR_TOKEN == 'true'
-        uses: docker/login-action@v4.0.0
+        uses: docker/login-action@v4.1.0
         with:
           registry: ghcr.io
           username: ${{ env.GHCR_OWNER }}

.github/workflows/pr-checklist-check.yml

@@ -1,45 +0,0 @@
-name: PR Checklist Check
-
-on:
-  pull_request_target:
-    types: [opened, edited, reopened, synchronize]
-
-jobs:
-  check:
-    runs-on: ubuntu-latest
-
-    permissions:
-      pull-requests: write
-      issues: write
-
-    steps:
-      - name: Check checklist
-        id: check
-        uses: actions/github-script@v7
-        with:
-          script: |
-            const body = context.payload.pull_request.body || "";
-            const regex = /-\s*\[\s*x\s*\].*没有.*认真阅读/i;
-            const bad = regex.test(body);
-            core.setOutput("bad", bad);
-
-      - name: Close PR
-        if: steps.check.outputs.bad == 'true'
-        uses: actions/github-script@v7
-        with:
-          script: |
-            const pr = context.payload.pull_request;
-
-            await github.rest.issues.createComment({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              issue_number: pr.number,
-              body: `检测到你勾选了“我没有认真阅读”，PR 已关闭。`
-            });
-
-            await github.rest.pulls.update({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              pull_number: pr.number,
-              state: "closed"
-            });

.github/workflows/pr-title-check.yml

@@ -0,0 +1,54 @@
+name: PR Title Check
+
+on:
+  pull_request_target:
+    types: [opened, edited, reopened, synchronize]
+
+jobs:
+  title-format:
+    if: github.repository == 'AstrBotDevs/AstrBot'
+    runs-on: ubuntu-latest
+    permissions:
+      pull-requests: write
+      issues: write
+
+    steps:
+      - name: Validate PR title
+        uses: actions/github-script@v8
+        with:
+          script: |
+            const title = (context.payload.pull_request.title || "").trim();
+            // allow only:
+            // feat: xxx
+            // feat(scope): xxx
+            const pattern = /^(feat)(\([a-z0-9-]+\))?:\s.+$/i;
+            const isValid = pattern.test(title);
+            const isSameRepo =
+              context.payload.pull_request.head.repo.full_name === context.payload.repository.full_name;
+
+            if (!isValid) {
+              if (isSameRepo) {
+                try {
+                  await github.rest.issues.createComment({
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    issue_number: context.payload.pull_request.number,
+                    body: [
+                      "⚠️ PR title format check failed.",
+                      "Required formats:",
+                      "- `feat: xxx`",
+                      "- `feat(scope): xxx`",
+                      "Please update your PR title and push again."
+                    ].join("\n")
+                  });
+                } catch (e) {
+                  core.warning(`Failed to post PR title comment: ${e.message}`);
+                }
+              } else {
+                core.warning("Fork PR: comment permission is restricted; skip posting review comment.");
+              }
+            }
+
+            if (!isValid) {
+              core.setFailed("Invalid PR title. Expected format: feat: xxx or feat(scope): xxx.");
+            }

.github/workflows/release.yml

@@ -20,6 +20,7 @@ permissions:
 jobs:
   build-dashboard:
     name: Build Dashboard
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-24.04
     env:
       R2_ACCOUNT_ID: ${{ secrets.R2_ACCOUNT_ID }}
@@ -50,7 +51,7 @@ jobs:
           echo "tag=$tag" >> "$GITHUB_OUTPUT"
 
       - name: Setup pnpm
-        uses: pnpm/action-setup@v4.4.0
+        uses: pnpm/action-setup@v5.0.0
         with:
           version: 10.28.2
 
@@ -104,6 +105,7 @@ jobs:
 
   publish-release:
     name: Publish GitHub Release
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-24.04
     needs:
       - build-dashboard
@@ -183,6 +185,7 @@ jobs:
 
   publish-pypi:
     name: Publish PyPI
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-24.04
     needs:
       - publish-release

.github/workflows/stale.yml

@@ -18,6 +18,7 @@ on:
 
 jobs:
   stale:
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-latest
     permissions:
       issues: write

.github/workflows/sync-wiki.yml

@@ -18,6 +18,7 @@ concurrency:
 
 jobs:
   sync:
+    if: github.repository == 'AstrBotDevs/AstrBot'
     runs-on: ubuntu-latest
     permissions:
       contents: read

.github/workflows/unit_tests.yml

@@ -0,0 +1,37 @@
+name: Unit Tests
+
+on:
+  push:
+    branches:
+      - master
+    paths-ignore:
+      - 'README*.md'
+      - 'changelogs/**'
+      - 'dashboard/**'
+  pull_request:
+  workflow_dispatch:
+
+jobs:
+  unit-tests:
+    name: Run pytest suite
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: '3.12'
+
+      - name: Install uv
+        run: |
+          python -m pip install --upgrade pip
+          python -m pip install uv
+
+      - name: Run tests
+        run: |
+          chmod +x scripts/run_pytests_ci.sh
+          bash ./scripts/run_pytests_ci.sh ./tests

.gitignore

@@ -63,3 +63,4 @@ GenieData/
 .kilocode/
 .worktrees/
 
+dashboard/bun.lock

Dockerfile

@@ -12,9 +12,11 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     ca-certificates \
     bash \
     ffmpeg \
+    libavcodec-extra \
     curl \
     gnupg \
     git \
+    ripgrep \
     && curl -fsSL https://deb.nodesource.com/setup_lts.x | bash - \
     && apt-get install -y --no-install-recommends nodejs \
     && apt-get clean \

README.md

@@ -1,4 +1,5 @@
-![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)
+![astrbot-github-banner-v2-light-0405_副本](https://github.com/user-attachments/assets/36fb04e4-cc75-4454-bd8b-049d11aa86f9)
+
 
 <div align="center">
 
@@ -32,7 +33,7 @@
 <a href="https://astrbot.app/">Documentation</a> ｜
 <a href="https://blog.astrbot.app/">Blog</a> ｜
 <a href="https://astrbot.featurebase.app/roadmap">Roadmap</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/issues">Issue Tracker</a>
+<a href="https://github.com/AstrBotDevs/AstrBot/issues">Issue Tracker</a> ｜
 <a href="mailto:community@astrbot.app">Email Support</a>
 </div>
 
@@ -92,6 +93,9 @@ Update `astrbot`:
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot deployed via `uv` **does not support upgrading through the WebUI**. To update, please run the command above from the command line.
+
 ### Docker Deployment
 
 For users familiar with containers and looking for a more stable, production-ready deployment method, we recommend deploying AstrBot with Docker / Docker Compose.
@@ -154,6 +158,7 @@ Connect AstrBot to your favorite chat platform.
 | LINE | Official |
 | Satori | Official |
 | Misskey | Official |
+| Mattermost | Official |
 | WhatsApp (Coming Soon) | Official |
 | [Matrix](https://github.com/stevessr/astrbot_plugin_matrix_adapter) | Community |
 | [KOOK](https://github.com/wuyan1003/astrbot_plugin_kook_adapter) | Community |
@@ -184,6 +189,7 @@ Connect AstrBot to your favorite chat platform.
 | Coze | LLMOps Platforms |
 | OpenAI Whisper | Speech-to-Text Services |
 | SenseVoice | Speech-to-Text Services |
+| Xiaomi MiMo Omni | Speech-to-Text Services |
 | OpenAI TTS | Text-to-Speech Services |
 | Gemini TTS | Text-to-Speech Services |
 | GPT-Sovits-Inference | Text-to-Speech Services |
@@ -193,6 +199,7 @@ Connect AstrBot to your favorite chat platform.
 | Alibaba Cloud Bailian TTS | Text-to-Speech Services |
 | Azure TTS | Text-to-Speech Services |
 | Minimax TTS | Text-to-Speech Services |
+| Xiaomi MiMo TTS | Text-to-Speech Services |
 | Volcano Engine TTS | Text-to-Speech Services |
 
 ## ❤️ Sponsors
@@ -225,14 +232,17 @@ pre-commit install
 
 ### QQ Groups
 
-- Group 9: 1076659624 (New)
-- Group 10: 1078079676 (New)
-- Group 1: 322154837
-- Group 3: 630166526
-- Group 5: 822130018
-- Group 6: 753075035
-- Group 7: 743746109
-- Group 8: 1030353265
+- Group 12: 916228568 (New)
+- Group 9: 1076659624 (Full)
+- Group 10: 1078079676 (Full)
+- Group 11: 704659519 (Full)
+- Group 1: 322154837 (Full)
+- Group 3: 630166526 (Full)
+- Group 4: 1077826412 (Full)
+- Group 5: 822130018 (Full)
+- Group 6: 753075035 (Full)
+- Group 7: 743746109 (Full)
+- Group 8: 1030353265 (Full)
 
 - Developer Group(Chit-chat): 975206796
 - Developer Group(Formal): 1039761811

README_fr.md

@@ -92,6 +92,9 @@ Mettre à jour `astrbot` :
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot déployé via `uv` **ne prend pas en charge la mise à jour via le WebUI**. Pour mettre à jour, exécutez la commande ci-dessus depuis le terminal.
+
 ### Déploiement Docker
 
 Pour les utilisateurs familiers avec les conteneurs et qui souhaitent une méthode plus stable et adaptée à la production, nous recommandons de déployer AstrBot avec Docker / Docker Compose.
@@ -184,6 +187,7 @@ Connectez AstrBot à vos plateformes de chat préférées.
 | Coze | Plateformes LLMOps |
 | OpenAI Whisper | Services de reconnaissance vocale |
 | SenseVoice | Services de reconnaissance vocale |
+| Xiaomi MiMo Omni | Services de reconnaissance vocale |
 | OpenAI TTS | Services de synthèse vocale |
 | Gemini TTS | Services de synthèse vocale |
 | GPT-Sovits-Inference | Services de synthèse vocale |
@@ -193,6 +197,7 @@ Connectez AstrBot à vos plateformes de chat préférées.
 | Alibaba Cloud Bailian TTS | Services de synthèse vocale |
 | Azure TTS | Services de synthèse vocale |
 | Minimax TTS | Services de synthèse vocale |
+| Xiaomi MiMo TTS | Services de synthèse vocale |
 | Volcano Engine TTS | Services de synthèse vocale |
 
 ## ❤️ Contribuer
@@ -217,10 +222,17 @@ pre-commit install
 
 ### Groupes QQ
 
-- Groupe 1 : 322154837
-- Groupe 3 : 630166526
-- Groupe 5 : 822130018
-- Groupe 6 : 753075035
+- Groupe 12 : 916228568 (nouveau)
+- Groupe 9 : 1076659624 (complet)
+- Groupe 10 : 1078079676 (complet)
+- Groupe 11 : 704659519 (complet)
+- Groupe 1 : 322154837 (complet)
+- Groupe 3 : 630166526 (complet)
+- Groupe 4 : 1077826412 (complet)
+- Groupe 5 : 822130018 (complet)
+- Groupe 6 : 753075035 (complet)
+- Groupe 7 : 743746109 (complet)
+- Groupe 8 : 1030353265 (complet)
 - Groupe développeurs : 975206796
 - Groupe développeurs (officiel) : 1039761811
 

README_ja.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> `uv` 経由でデプロイした AstrBot は、**WebUI からのバージョンアップグレードに対応していません**。更新するには、上記のコマンドをコマンドラインで実行してください。
+
 ### Docker デプロイ
 
 コンテナ運用に慣れており、より安定した本番向けのデプロイ方法を求めるユーザーには、Docker / Docker Compose での AstrBot デプロイをおすすめします。
@@ -185,6 +188,7 @@ AstrBot をよく使うチャットプラットフォームに接続できます
 | Coze | LLMOps プラットフォーム |
 | OpenAI Whisper | 音声認識サービス |
 | SenseVoice | 音声認識サービス |
+| Xiaomi MiMo Omni | 音声認識サービス |
 | OpenAI TTS | 音声合成サービス |
 | Gemini TTS | 音声合成サービス |
 | GPT-Sovits-Inference | 音声合成サービス |
@@ -194,6 +198,7 @@ AstrBot をよく使うチャットプラットフォームに接続できます
 | Alibaba Cloud 百炼 TTS | 音声合成サービス |
 | Azure TTS | 音声合成サービス |
 | Minimax TTS | 音声合成サービス |
+| Xiaomi MiMo TTS | 音声合成サービス |
 | Volcano Engine TTS | 音声合成サービス |
 
 ## ❤️ コントリビューション
@@ -218,10 +223,17 @@ pre-commit install
 
 ### QQ グループ
 
-- 1群: 322154837
-- 3群: 630166526
-- 5群: 822130018
-- 6群: 753075035
+- 12群: 916228568 (新)
+- 9群: 1076659624 (満員)
+- 10群: 1078079676 (満員)
+- 11群: 704659519 (満員)
+- 1群: 322154837 (満員)
+- 3群: 630166526 (満員)
+- 4群: 1077826412 (満員)
+- 5群: 822130018 (満員)
+- 6群: 753075035 (満員)
+- 7群: 743746109 (満員)
+- 8群: 1030353265 (満員)
 - 開発者群: 975206796
 - 開発者群（正式）: 1039761811
 

README_ru.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot, развёрнутый через `uv`, **не поддерживает обновление через WebUI**. Для обновления выполните указанную выше команду из командной строки.
+
 ### Развёртывание Docker
 
 Для пользователей, знакомых с контейнерами и которым нужен более стабильный и подходящий для production способ, мы рекомендуем разворачивать AstrBot через Docker / Docker Compose.
@@ -184,6 +187,7 @@ yay -S astrbot-git
 | Coze | Платформы LLMOps |
 | OpenAI Whisper | Сервисы распознавания речи |
 | SenseVoice | Сервисы распознавания речи |
+| Xiaomi MiMo Omni | Сервисы распознавания речи |
 | OpenAI TTS | Сервисы синтеза речи |
 | Gemini TTS | Сервисы синтеза речи |
 | GPT-Sovits-Inference | Сервисы синтеза речи |
@@ -193,6 +197,7 @@ yay -S astrbot-git
 | Alibaba Cloud Bailian TTS | Сервисы синтеза речи |
 | Azure TTS | Сервисы синтеза речи |
 | Minimax TTS | Сервисы синтеза речи |
+| Xiaomi MiMo TTS | Сервисы синтеза речи |
 | Volcano Engine TTS | Сервисы синтеза речи |
 
 ## ❤️ Вклад в проект
@@ -217,10 +222,17 @@ pre-commit install
 
 ### Группы QQ
 
-- Группа 1: 322154837
-- Группа 3: 630166526
-- Группа 5: 822130018
-- Группа 6: 753075035
+- Группа 12: 916228568 (новая)
+- Группа 9: 1076659624 (полная)
+- Группа 10: 1078079676 (полная)
+- Группа 11: 704659519 (полная)
+- Группа 1: 322154837 (полная)
+- Группа 3: 630166526 (полная)
+- Группа 4: 1077826412 (полная)
+- Группа 5: 822130018 (полная)
+- Группа 6: 753075035 (полная)
+- Группа 7: 743746109 (полная)
+- Группа 8: 1030353265 (полная)
 - Группа разработчиков: 975206796
 - Группа разработчиков (официальная): 1039761811
 

README_zh-TW.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> 透過 `uv` 部署的 AstrBot **不支援在 WebUI 中進行版本升級**。如需更新，請透過命令列執行上述命令。
+
 ### Docker 部署
 
 對於熟悉容器、希望獲得更穩定且更適合正式環境部署方式的使用者，我們推薦使用 Docker / Docker Compose 部署 AstrBot。
@@ -184,6 +187,7 @@ yay -S astrbot-git
 | Coze | LLMOps 平台 |
 | OpenAI Whisper | 語音轉文字服務 |
 | SenseVoice | 語音轉文字服務 |
+| Xiaomi MiMo Omni | 語音轉文字服務 |
 | OpenAI TTS | 文字轉語音服務 |
 | Gemini TTS | 文字轉語音服務 |
 | GPT-Sovits-Inference | 文字轉語音服務 |
@@ -193,6 +197,7 @@ yay -S astrbot-git
 | 阿里雲百煉 TTS | 文字轉語音服務 |
 | Azure TTS | 文字轉語音服務 |
 | Minimax TTS | 文字轉語音服務 |
+| Xiaomi MiMo TTS | 文字轉語音服務 |
 | 火山引擎 TTS | 文字轉語音服務 |
 
 ## ❤️ 貢獻
@@ -217,14 +222,17 @@ pre-commit install
 
 ### QQ 群組
 
-- 9 群: 1076659624 (新)
-- 10 群: 1078079676 (新)
-- 1 群：322154837
-- 3 群：630166526
-- 5 群：822130018
-- 6 群：753075035
-- 7 群：743746109
-- 8 群：1030353265
+- 12 群：916228568 (新)
+- 9 群：1076659624 (人滿)
+- 10 群：1078079676 (人滿)
+- 11 群：704659519 (人滿)
+- 1 群：322154837 (人滿)
+- 3 群：630166526 (人滿)
+- 4 群：1077826412 (人滿)
+- 5 群：822130018 (人滿)
+- 6 群：753075035 (人滿)
+- 7 群：743746109 (人滿)
+- 8 群：1030353265 (人滿)
 - 開發者群（闲聊吹水）：975206796
 - 開發者群（正式）：1039761811
 

README_zh.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> 通过 `uv` 部署的 AstrBot **不支持在 WebUI 中进行版本升级**。如需更新，请通过命令行执行上述命令。
+
 ### Docker 部署
 
 对于熟悉容器、希望获得更稳定且更适合生产环境部署方式的用户，我们推荐使用 Docker / Docker Compose 部署 AstrBot。
@@ -185,6 +188,7 @@ yay -S astrbot-git
 | Coze | LLMOps 平台 |
 | OpenAI Whisper | 语音转文本 |
 | SenseVoice | 语音转文本 |
+| Xiaomi MiMo Omni | 语音转文本 |
 | OpenAI TTS | 文本转语音 |
 | Gemini TTS | 文本转语音 |
 | GPT-Sovits-Inference | 文本转语音 |
@@ -194,6 +198,7 @@ yay -S astrbot-git
 | 阿里云百炼 TTS | 文本转语音 |
 | Azure TTS | 文本转语音 |
 | Minimax TTS | 文本转语音 |
+| Xiaomi MiMo TTS | 文本转语音 |
 | 火山引擎 TTS | 文本转语音 |
 
 ## ❤️ 贡献
@@ -218,14 +223,17 @@ pre-commit install
 
 ### QQ 群组
 
-- 9 群: 1076659624 (新)
-- 10 群: 1078079676 (新)
-- 1 群：322154837
-- 3 群：630166526
-- 5 群：822130018
-- 6 群：753075035
-- 7 群：743746109
-- 8 群：1030353265
+- 12 群：916228568 (新)
+- 9 群：1076659624 (人满)
+- 10 群：1078079676 (人满)
+- 11 群：704659519 (人满)
+- 1 群：322154837 (人满)
+- 3 群：630166526 (人满)
+- 4 群：1077826412 (人满)
+- 5 群：822130018 (人满)
+- 6 群：753075035 (人满)
+- 7 群：743746109 (人满)
+- 8 群：1030353265 (人满)
 - 开发者群（偏闲聊吹水）：975206796
 - 开发者群（正式）：1039761811
 

astrbot/builtin_stars/astrbot/main.py

@@ -36,9 +36,9 @@ class Main(star.Star):
         if self.ltm_enabled(event) and self.ltm and has_image_or_plain:
             need_active = await self.ltm.need_active_reply(event)
 
-            group_icl_enable = self.context.get_config()["provider_ltm_settings"][
-                "group_icl_enable"
-            ]
+            group_icl_enable = self.context.get_config(umo=event.unified_msg_origin)[
+                "provider_ltm_settings"
+            ]["group_icl_enable"]
             if group_icl_enable:
                 """记录对话"""
                 try:

astrbot/builtin_stars/builtin_commands/commands/__init__.py

@@ -1,29 +1,15 @@
 # Commands module
 
 from .admin import AdminCommands
-from .alter_cmd import AlterCmdCommands
 from .conversation import ConversationCommands
 from .help import HelpCommand
-from .llm import LLMCommands
-from .persona import PersonaCommands
-from .plugin import PluginCommands
-from .provider import ProviderCommands
 from .setunset import SetUnsetCommands
 from .sid import SIDCommand
-from .t2i import T2ICommand
-from .tts import TTSCommand
 
 __all__ = [
     "AdminCommands",
-    "AlterCmdCommands",
     "ConversationCommands",
     "HelpCommand",
-    "LLMCommands",
-    "PersonaCommands",
-    "PluginCommands",
-    "ProviderCommands",
-    "SIDCommand",
     "SetUnsetCommands",
-    "T2ICommand",
-    "TTSCommand",
+    "SIDCommand",
 ]

astrbot/builtin_stars/builtin_commands/commands/admin.py

@@ -1,77 +1,17 @@
 from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageChain, MessageEventResult
+from astrbot.api.event import AstrMessageEvent, MessageChain
 from astrbot.core.config.default import VERSION
 from astrbot.core.utils.io import download_dashboard
 
+from ..i18n import t
+
 
 class AdminCommands:
     def __init__(self, context: star.Context) -> None:
         self.context = context
 
-    async def op(self, event: AstrMessageEvent, admin_id: str = "") -> None:
-        """授权管理员。op <admin_id>"""
-        if not admin_id:
-            event.set_result(
-                MessageEventResult().message(
-                    "使用方法: /op <id> 授权管理员；/deop <id> 取消管理员。可通过 /sid 获取 ID。",
-                ),
-            )
-            return
-        self.context.get_config()["admins_id"].append(str(admin_id))
-        self.context.get_config().save_config()
-        event.set_result(MessageEventResult().message("授权成功。"))
-
-    async def deop(self, event: AstrMessageEvent, admin_id: str = "") -> None:
-        """取消授权管理员。deop <admin_id>"""
-        if not admin_id:
-            event.set_result(
-                MessageEventResult().message(
-                    "使用方法: /deop <id> 取消管理员。可通过 /sid 获取 ID。",
-                ),
-            )
-            return
-        try:
-            self.context.get_config()["admins_id"].remove(str(admin_id))
-            self.context.get_config().save_config()
-            event.set_result(MessageEventResult().message("取消授权成功。"))
-        except ValueError:
-            event.set_result(
-                MessageEventResult().message("此用户 ID 不在管理员名单内。"),
-            )
-
-    async def wl(self, event: AstrMessageEvent, sid: str = "") -> None:
-        """添加白名单。wl <sid>"""
-        if not sid:
-            event.set_result(
-                MessageEventResult().message(
-                    "使用方法: /wl <id> 添加白名单；/dwl <id> 删除白名单。可通过 /sid 获取 ID。",
-                ),
-            )
-            return
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        cfg["platform_settings"]["id_whitelist"].append(str(sid))
-        cfg.save_config()
-        event.set_result(MessageEventResult().message("添加白名单成功。"))
-
-    async def dwl(self, event: AstrMessageEvent, sid: str = "") -> None:
-        """删除白名单。dwl <sid>"""
-        if not sid:
-            event.set_result(
-                MessageEventResult().message(
-                    "使用方法: /dwl <id> 删除白名单。可通过 /sid 获取 ID。",
-                ),
-            )
-            return
-        try:
-            cfg = self.context.get_config(umo=event.unified_msg_origin)
-            cfg["platform_settings"]["id_whitelist"].remove(str(sid))
-            cfg.save_config()
-            event.set_result(MessageEventResult().message("删除白名单成功。"))
-        except ValueError:
-            event.set_result(MessageEventResult().message("此 SID 不在白名单内。"))
-
     async def update_dashboard(self, event: AstrMessageEvent) -> None:
         """更新管理面板"""
-        await event.send(MessageChain().message("正在尝试更新管理面板..."))
+        await event.send(MessageChain().message(t(self.context, "dashboard.updating")))
         await download_dashboard(version=f"v{VERSION}", latest=False)
-        await event.send(MessageChain().message("管理面板更新完成。"))
+        await event.send(MessageChain().message(t(self.context, "dashboard.updated")))

astrbot/builtin_stars/builtin_commands/commands/alter_cmd.py

@@ -1,173 +0,0 @@
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageChain
-from astrbot.core.star.filter.command import CommandFilter
-from astrbot.core.star.filter.command_group import CommandGroupFilter
-from astrbot.core.star.filter.permission import PermissionTypeFilter
-from astrbot.core.star.star import star_map
-from astrbot.core.star.star_handler import StarHandlerMetadata, star_handlers_registry
-from astrbot.core.utils.command_parser import CommandParserMixin
-
-from .utils.rst_scene import RstScene
-
-
-class AlterCmdCommands(CommandParserMixin):
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    async def update_reset_permission(self, scene_key: str, perm_type: str) -> None:
-        """更新reset命令在特定场景下的权限设置"""
-        from astrbot.api import sp
-
-        alter_cmd_cfg = await sp.global_get("alter_cmd", {})
-        plugin_cfg = alter_cmd_cfg.get("astrbot", {})
-        reset_cfg = plugin_cfg.get("reset", {})
-        reset_cfg[scene_key] = perm_type
-        plugin_cfg["reset"] = reset_cfg
-        alter_cmd_cfg["astrbot"] = plugin_cfg
-        await sp.global_put("alter_cmd", alter_cmd_cfg)
-
-    async def alter_cmd(self, event: AstrMessageEvent) -> None:
-        token = self.parse_commands(event.message_str)
-        if token.len < 3:
-            await event.send(
-                MessageChain().message(
-                    "该指令用于设置指令或指令组的权限。\n"
-                    "格式: /alter_cmd <cmd_name> <admin/member>\n"
-                    "例1: /alter_cmd c1 admin 将 c1 设为管理员指令\n"
-                    "例2: /alter_cmd g1 c1 admin 将 g1 指令组的 c1 子指令设为管理员指令\n"
-                    "/alter_cmd reset config 打开 reset 权限配置",
-                ),
-            )
-            return
-
-        # 兼容 reset scene 的专门配置
-        cmd_name = token.get(1)
-        cmd_type = token.get(2)
-
-        if cmd_name == "reset" and cmd_type == "config":
-            from astrbot.api import sp
-
-            alter_cmd_cfg = await sp.global_get("alter_cmd", {})
-            plugin_ = alter_cmd_cfg.get("astrbot", {})
-            reset_cfg = plugin_.get("reset", {})
-
-            group_unique_on = reset_cfg.get("group_unique_on", "admin")
-            group_unique_off = reset_cfg.get("group_unique_off", "admin")
-            private = reset_cfg.get("private", "member")
-
-            config_menu = f"""reset命令权限细粒度配置
-                当前配置：
-                1. 群聊+会话隔离开: {group_unique_on}
-                2. 群聊+会话隔离关: {group_unique_off}
-                3. 私聊: {private}
-                修改指令格式：
-                /alter_cmd reset scene <场景编号> <admin/member>
-                例如: /alter_cmd reset scene 2 member"""
-            await event.send(MessageChain().message(config_menu))
-            return
-
-        if cmd_name == "reset" and cmd_type == "scene" and token.len >= 4:
-            scene_num = token.get(3)
-            perm_type = token.get(4)
-
-            if scene_num is None or perm_type is None:
-                await event.send(MessageChain().message("场景编号和权限类型不能为空"))
-                return
-
-            if not scene_num.isdigit() or int(scene_num) < 1 or int(scene_num) > 3:
-                await event.send(
-                    MessageChain().message("场景编号必须是 1-3 之间的数字"),
-                )
-                return
-
-            if perm_type not in ["admin", "member"]:
-                await event.send(
-                    MessageChain().message("权限类型错误，只能是 admin 或 member"),
-                )
-                return
-
-            scene_num = int(scene_num)
-            scene = RstScene.from_index(scene_num)
-            scene_key = scene.key
-
-            await self.update_reset_permission(scene_key, perm_type)
-
-            await event.send(
-                MessageChain().message(
-                    f"已将 reset 命令在{scene.name}场景下的权限设为{perm_type}",
-                ),
-            )
-            return
-
-        if cmd_type not in ["admin", "member"]:
-            await event.send(
-                MessageChain().message("指令类型错误，可选类型有 admin, member"),
-            )
-            return
-
-        # 查找指令
-        cmd_name = " ".join(token.tokens[1:-1])
-        cmd_type = token.get(-1)
-        found_command = None
-        cmd_group = False
-        for handler in star_handlers_registry:
-            assert isinstance(handler, StarHandlerMetadata)
-            for filter_ in handler.event_filters:
-                if isinstance(filter_, CommandFilter):
-                    if filter_.equals(cmd_name):
-                        found_command = handler
-                        break
-                elif isinstance(filter_, CommandGroupFilter):
-                    if filter_.equals(cmd_name):
-                        found_command = handler
-                        cmd_group = True
-                        break
-
-        if not found_command:
-            await event.send(MessageChain().message("未找到该指令"))
-            return
-
-        found_plugin = star_map[found_command.handler_module_path]
-
-        from astrbot.api import sp
-
-        alter_cmd_cfg = await sp.global_get("alter_cmd", {})
-        plugin_ = alter_cmd_cfg.get(found_plugin.name, {})
-        cfg = plugin_.get(found_command.handler_name, {})
-        cfg["permission"] = cmd_type
-        plugin_[found_command.handler_name] = cfg
-        alter_cmd_cfg[found_plugin.name] = plugin_
-
-        await sp.global_put("alter_cmd", alter_cmd_cfg)
-
-        # 注入权限过滤器
-        found_permission_filter = False
-        for filter_ in found_command.event_filters:
-            if isinstance(filter_, PermissionTypeFilter):
-                if cmd_type == "admin":
-                    from astrbot.api.event import filter
-
-                    filter_.permission_type = filter.PermissionType.ADMIN
-                else:
-                    from astrbot.api.event import filter
-
-                    filter_.permission_type = filter.PermissionType.MEMBER
-                found_permission_filter = True
-                break
-        if not found_permission_filter:
-            from astrbot.api.event import filter
-
-            found_command.event_filters.insert(
-                0,
-                PermissionTypeFilter(
-                    filter.PermissionType.ADMIN
-                    if cmd_type == "admin"
-                    else filter.PermissionType.MEMBER,
-                ),
-            )
-        cmd_group_str = "指令组" if cmd_group else "指令"
-        await event.send(
-            MessageChain().message(
-                f"已将「{cmd_name}」{cmd_group_str} 的权限级别调整为 {cmd_type}。",
-            ),
-        )

astrbot/builtin_stars/builtin_commands/commands/conversation.py

@@ -1,15 +1,15 @@
-import datetime
-
 from astrbot.api import sp, star
 from astrbot.api.event import AstrMessageEvent, MessageEventResult
+from astrbot.core import logger
 from astrbot.core.agent.runners.deerflow.constants import (
+    DEERFLOW_AGENT_RUNNER_PROVIDER_ID_KEY,
     DEERFLOW_PROVIDER_TYPE,
     DEERFLOW_THREAD_ID_KEY,
 )
-from astrbot.core.platform.astr_message_event import MessageSession
-from astrbot.core.platform.message_type import MessageType
+from astrbot.core.agent.runners.deerflow.deerflow_api_client import DeerFlowAPIClient
 from astrbot.core.utils.active_event_registry import active_event_registry
 
+from ..i18n import t
 from .utils.rst_scene import RstScene
 
 THIRD_PARTY_AGENT_RUNNER_KEY = {
@@ -21,6 +21,85 @@ THIRD_PARTY_AGENT_RUNNER_KEY = {
 THIRD_PARTY_AGENT_RUNNER_STR = ", ".join(THIRD_PARTY_AGENT_RUNNER_KEY.keys())
 
 
+async def _cleanup_deerflow_thread_if_present(
+    context: star.Context,
+    umo: str,
+) -> None:
+    try:
+        thread_id = await sp.get_async(
+            scope="umo",
+            scope_id=umo,
+            key=DEERFLOW_THREAD_ID_KEY,
+            default="",
+        )
+        if not thread_id:
+            return
+
+        cfg = context.get_config(umo=umo)
+        provider_id = cfg["provider_settings"].get(
+            DEERFLOW_AGENT_RUNNER_PROVIDER_ID_KEY,
+            "",
+        )
+        if not provider_id:
+            return
+
+        merged_provider_config = context.provider_manager.get_provider_config_by_id(
+            provider_id,
+            merged=True,
+        )
+        if not merged_provider_config:
+            logger.warning(
+                "Failed to resolve DeerFlow provider config for remote thread cleanup: provider_id=%s",
+                provider_id,
+            )
+            return
+
+        client = DeerFlowAPIClient(
+            api_base=merged_provider_config.get(
+                "deerflow_api_base",
+                "http://127.0.0.1:2026",
+            ),
+            api_key=merged_provider_config.get("deerflow_api_key", ""),
+            auth_header=merged_provider_config.get("deerflow_auth_header", ""),
+            proxy=merged_provider_config.get("proxy", ""),
+        )
+        try:
+            await client.delete_thread(thread_id)
+        finally:
+            try:
+                await client.close()
+            except Exception as e:
+                logger.warning(
+                    "Failed to close DeerFlow API client after thread cleanup: %s",
+                    e,
+                )
+    except Exception as e:
+        logger.warning(
+            "Failed to clean up DeerFlow thread for session %s: %s",
+            umo,
+            e,
+        )
+
+
+async def _clear_third_party_agent_runner_state(
+    context: star.Context,
+    umo: str,
+    agent_runner_type: str,
+) -> None:
+    session_key = THIRD_PARTY_AGENT_RUNNER_KEY.get(agent_runner_type)
+    if not session_key:
+        return
+
+    if agent_runner_type == DEERFLOW_PROVIDER_TYPE:
+        await _cleanup_deerflow_thread_if_present(context, umo)
+
+    await sp.remove_async(
+        scope="umo",
+        scope_id=umo,
+        key=session_key,
+    )
+
+
 class ConversationCommands:
     def __init__(self, context: star.Context) -> None:
         self.context = context
@@ -60,8 +139,12 @@ class ConversationCommands:
         if required_perm == "admin" and message.role != "admin":
             message.set_result(
                 MessageEventResult().message(
-                    f"在{scene.name}场景下，reset命令需要管理员权限，"
-                    f"您 (ID {message.get_sender_id()}) 不是管理员，无法执行此操作。",
+                    t(
+                        self.context,
+                        "conversation.reset_admin_required",
+                        scene_name=t(self.context, f"scene.{scene.key}"),
+                        sender_id=message.get_sender_id(),
+                    ),
                 ),
             )
             return
@@ -69,17 +152,23 @@ class ConversationCommands:
         agent_runner_type = cfg["provider_settings"]["agent_runner_type"]
         if agent_runner_type in THIRD_PARTY_AGENT_RUNNER_KEY:
             active_event_registry.stop_all(umo, exclude=message)
-            await sp.remove_async(
-                scope="umo",
-                scope_id=umo,
-                key=THIRD_PARTY_AGENT_RUNNER_KEY[agent_runner_type],
+            await _clear_third_party_agent_runner_state(
+                self.context,
+                umo,
+                agent_runner_type,
+            )
+            message.set_result(
+                MessageEventResult().message(
+                    t(self.context, "conversation.reset_success"),
+                )
             )
-            message.set_result(MessageEventResult().message("重置对话成功。"))
             return
 
         if not self.context.get_using_provider(umo):
             message.set_result(
-                MessageEventResult().message("未找到任何 LLM 提供商。请先配置。"),
+                MessageEventResult().message(
+                    t(self.context, "conversation.no_provider"),
+                ),
             )
             return
 
@@ -88,7 +177,7 @@ class ConversationCommands:
         if not cid:
             message.set_result(
                 MessageEventResult().message(
-                    "当前未处于对话状态，请 /switch 切换或者 /new 创建。",
+                    t(self.context, "conversation.no_conversation"),
                 ),
             )
             return
@@ -101,7 +190,7 @@ class ConversationCommands:
             [],
         )
 
-        ret = "清除聊天历史成功！"
+        ret = t(self.context, "conversation.reset_success")
 
         message.set_extra("_clean_ltm_session", True)
 
@@ -124,160 +213,37 @@ class ConversationCommands:
         if stopped_count > 0:
             message.set_result(
                 MessageEventResult().message(
-                    f"已请求停止 {stopped_count} 个运行中的任务。"
+                    t(
+                        self.context,
+                        "conversation.stop_requested",
+                        count=stopped_count,
+                    ),
                 )
             )
             return
 
-        message.set_result(MessageEventResult().message("当前会话没有运行中的任务。"))
-
-    async def his(self, message: AstrMessageEvent, page: int = 1) -> None:
-        """查看对话记录"""
-        if not self.context.get_using_provider(message.unified_msg_origin):
-            message.set_result(
-                MessageEventResult().message("未找到任何 LLM 提供商。请先配置。"),
+        message.set_result(
+            MessageEventResult().message(
+                t(self.context, "conversation.no_running_tasks"),
             )
-            return
-
-        size_per_page = 6
-
-        conv_mgr = self.context.conversation_manager
-        umo = message.unified_msg_origin
-        session_curr_cid = await conv_mgr.get_curr_conversation_id(umo)
-
-        if not session_curr_cid:
-            session_curr_cid = await conv_mgr.new_conversation(
-                umo,
-                message.get_platform_id(),
-            )
-
-        contexts, total_pages = await conv_mgr.get_human_readable_context(
-            umo,
-            session_curr_cid,
-            page,
-            size_per_page,
         )
 
-        parts = []
-        for context in contexts:
-            if len(context) > 150:
-                context = context[:150] + "..."
-            parts.append(f"{context}\n")
-
-        history = "".join(parts)
-        ret = (
-            f"当前对话历史记录："
-            f"{history or '无历史记录'}\n\n"
-            f"第 {page} 页 | 共 {total_pages} 页\n"
-            f"*输入 /history 2 跳转到第 2 页"
-        )
-
-        message.set_result(MessageEventResult().message(ret).use_t2i(False))
-
-    async def convs(self, message: AstrMessageEvent, page: int = 1) -> None:
-        """查看对话列表"""
-        cfg = self.context.get_config(umo=message.unified_msg_origin)
-        agent_runner_type = cfg["provider_settings"]["agent_runner_type"]
-        if agent_runner_type in THIRD_PARTY_AGENT_RUNNER_KEY:
-            message.set_result(
-                MessageEventResult().message(
-                    f"{THIRD_PARTY_AGENT_RUNNER_STR} 对话列表功能暂不支持。",
-                ),
-            )
-            return
-
-        size_per_page = 6
-        """获取所有对话列表"""
-        conversations_all = await self.context.conversation_manager.get_conversations(
-            message.unified_msg_origin,
-        )
-        """计算总页数"""
-        total_pages = (len(conversations_all) + size_per_page - 1) // size_per_page
-        """确保页码有效"""
-        page = max(1, min(page, total_pages))
-        """分页处理"""
-        start_idx = (page - 1) * size_per_page
-        end_idx = start_idx + size_per_page
-        conversations_paged = conversations_all[start_idx:end_idx]
-
-        parts = ["对话列表：\n---\n"]
-        """全局序号从当前页的第一个开始"""
-        global_index = start_idx + 1
-
-        """生成所有对话的标题字典"""
-        _titles = {}
-        for conv in conversations_all:
-            title = conv.title if conv.title else "新对话"
-            _titles[conv.cid] = title
-
-        """遍历分页后的对话生成列表显示"""
-        provider_settings = cfg.get("provider_settings", {})
-        platform_name = message.get_platform_name()
-        for conv in conversations_paged:
-            (
-                persona_id,
-                _,
-                force_applied_persona_id,
-                _,
-            ) = await self.context.persona_manager.resolve_selected_persona(
-                umo=message.unified_msg_origin,
-                conversation_persona_id=conv.persona_id,
-                platform_name=platform_name,
-                provider_settings=provider_settings,
-            )
-            if persona_id == "[%None]":
-                persona_name = "无"
-            elif persona_id:
-                persona_name = persona_id
-            else:
-                persona_name = "无"
-
-            if force_applied_persona_id:
-                persona_name = f"{persona_name} (自定义规则)"
-
-            title = _titles.get(conv.cid, "新对话")
-            parts.append(
-                f"{global_index}. {title}({conv.cid[:4]})\n  人格情景: {persona_name}\n  上次更新: {datetime.datetime.fromtimestamp(conv.updated_at).strftime('%m-%d %H:%M')}\n"
-            )
-            global_index += 1
-
-        parts.append("---\n")
-        ret = "".join(parts)
-        curr_cid = await self.context.conversation_manager.get_curr_conversation_id(
-            message.unified_msg_origin,
-        )
-        if curr_cid:
-            """从所有对话的标题字典中获取标题"""
-            title = _titles.get(curr_cid, "新对话")
-            ret += f"\n当前对话: {title}({curr_cid[:4]})"
-        else:
-            ret += "\n当前对话: 无"
-
-        cfg = self.context.get_config(umo=message.unified_msg_origin)
-        unique_session = cfg["platform_settings"]["unique_session"]
-        if unique_session:
-            ret += "\n会话隔离粒度: 个人"
-        else:
-            ret += "\n会话隔离粒度: 群聊"
-
-        ret += f"\n第 {page} 页 | 共 {total_pages} 页"
-        ret += "\n*输入 /ls 2 跳转到第 2 页"
-
-        message.set_result(MessageEventResult().message(ret).use_t2i(False))
-        return
-
     async def new_conv(self, message: AstrMessageEvent) -> None:
         """创建新对话"""
         cfg = self.context.get_config(umo=message.unified_msg_origin)
         agent_runner_type = cfg["provider_settings"]["agent_runner_type"]
         if agent_runner_type in THIRD_PARTY_AGENT_RUNNER_KEY:
             active_event_registry.stop_all(message.unified_msg_origin, exclude=message)
-            await sp.remove_async(
-                scope="umo",
-                scope_id=message.unified_msg_origin,
-                key=THIRD_PARTY_AGENT_RUNNER_KEY[agent_runner_type],
+            await _clear_third_party_agent_runner_state(
+                self.context,
+                message.unified_msg_origin,
+                agent_runner_type,
+            )
+            message.set_result(
+                MessageEventResult().message(
+                    t(self.context, "conversation.new_created")
+                )
             )
-            message.set_result(MessageEventResult().message("已创建新对话。"))
             return
 
         active_event_registry.stop_all(message.unified_msg_origin, exclude=message)
@@ -291,130 +257,11 @@ class ConversationCommands:
         message.set_extra("_clean_ltm_session", True)
 
         message.set_result(
-            MessageEventResult().message(f"切换到新对话: 新对话({cid[:4]})。"),
+            MessageEventResult().message(
+                t(
+                    self.context,
+                    "conversation.switched_new",
+                    conversation_id=cid[:4],
+                ),
+            ),
         )
-
-    async def groupnew_conv(self, message: AstrMessageEvent, sid: str = "") -> None:
-        """创建新群聊对话"""
-        if sid:
-            session = str(
-                MessageSession(
-                    platform_name=message.platform_meta.id,
-                    message_type=MessageType("GroupMessage"),
-                    session_id=sid,
-                ),
-            )
-
-            cpersona = await self._get_current_persona_id(session)
-            cid = await self.context.conversation_manager.new_conversation(
-                session,
-                message.get_platform_id(),
-                persona_id=cpersona,
-            )
-            message.set_result(
-                MessageEventResult().message(
-                    f"群聊 {session} 已切换到新对话: 新对话({cid[:4]})。",
-                ),
-            )
-        else:
-            message.set_result(
-                MessageEventResult().message("请输入群聊 ID。/groupnew 群聊ID。"),
-            )
-
-    async def switch_conv(
-        self,
-        message: AstrMessageEvent,
-        index: int | None = None,
-    ) -> None:
-        """通过 /ls 前面的序号切换对话"""
-        if not isinstance(index, int):
-            message.set_result(
-                MessageEventResult().message("类型错误，请输入数字对话序号。"),
-            )
-            return
-
-        if index is None:
-            message.set_result(
-                MessageEventResult().message(
-                    "请输入对话序号。/switch 对话序号。/ls 查看对话 /new 新建对话",
-                ),
-            )
-            return
-        conversations = await self.context.conversation_manager.get_conversations(
-            message.unified_msg_origin,
-        )
-        if index > len(conversations) or index < 1:
-            message.set_result(
-                MessageEventResult().message("对话序号错误，请使用 /ls 查看"),
-            )
-        else:
-            conversation = conversations[index - 1]
-            title = conversation.title if conversation.title else "新对话"
-            await self.context.conversation_manager.switch_conversation(
-                message.unified_msg_origin,
-                conversation.cid,
-            )
-            message.set_result(
-                MessageEventResult().message(
-                    f"切换到对话: {title}({conversation.cid[:4]})。",
-                ),
-            )
-
-    async def rename_conv(self, message: AstrMessageEvent, new_name: str = "") -> None:
-        """重命名对话"""
-        if not new_name:
-            message.set_result(MessageEventResult().message("请输入新的对话名称。"))
-            return
-        await self.context.conversation_manager.update_conversation_title(
-            message.unified_msg_origin,
-            new_name,
-        )
-        message.set_result(MessageEventResult().message("重命名对话成功。"))
-
-    async def del_conv(self, message: AstrMessageEvent) -> None:
-        """删除当前对话"""
-        umo = message.unified_msg_origin
-        cfg = self.context.get_config(umo=umo)
-        is_unique_session = cfg["platform_settings"]["unique_session"]
-        if message.get_group_id() and not is_unique_session and message.role != "admin":
-            # 群聊，没开独立会话，发送人不是管理员
-            message.set_result(
-                MessageEventResult().message(
-                    f"会话处于群聊，并且未开启独立会话，并且您 (ID {message.get_sender_id()}) 不是管理员，因此没有权限删除当前对话。",
-                ),
-            )
-            return
-
-        agent_runner_type = cfg["provider_settings"]["agent_runner_type"]
-        if agent_runner_type in THIRD_PARTY_AGENT_RUNNER_KEY:
-            active_event_registry.stop_all(umo, exclude=message)
-            await sp.remove_async(
-                scope="umo",
-                scope_id=umo,
-                key=THIRD_PARTY_AGENT_RUNNER_KEY[agent_runner_type],
-            )
-            message.set_result(MessageEventResult().message("重置对话成功。"))
-            return
-
-        session_curr_cid = (
-            await self.context.conversation_manager.get_curr_conversation_id(umo)
-        )
-
-        if not session_curr_cid:
-            message.set_result(
-                MessageEventResult().message(
-                    "当前未处于对话状态，请 /switch 序号 切换或 /new 创建。",
-                ),
-            )
-            return
-
-        active_event_registry.stop_all(umo, exclude=message)
-
-        await self.context.conversation_manager.delete_conversation(
-            umo,
-            session_curr_cid,
-        )
-
-        ret = "删除当前对话成功。不再处于对话状态，使用 /switch 序号 切换到其他对话或 /new 创建。"
-        message.set_extra("_clean_ltm_session", True)
-        message.set_result(MessageEventResult().message(ret))

astrbot/builtin_stars/builtin_commands/commands/help.py

@@ -6,6 +6,8 @@ from astrbot.core.config.default import VERSION
 from astrbot.core.star import command_management
 from astrbot.core.utils.io import get_dashboard_version
 
+from ..i18n import t
+
 
 class HelpCommand:
     def __init__(self, context: star.Context) -> None:
@@ -32,7 +34,6 @@ class HelpCommand:
             return []
 
         lines: list[str] = []
-        hidden_commands = {"set", "unset", "websearch"}
 
         def walk(items: list[dict], indent: int = 0) -> None:
             for item in items:
@@ -49,9 +50,12 @@ class HelpCommand:
                     or item.get("original_command")
                     or item.get("handler_name")
                 )
-                if not effective:
-                    continue
-                if effective in hidden_commands:
+                if not effective or effective in [
+                    "set",
+                    "unset",
+                    "help",
+                    "dashboard_update",
+                ]:
                     continue
 
                 description = item.get("description") or ""
@@ -73,12 +77,13 @@ class HelpCommand:
         dashboard_version = await get_dashboard_version()
         command_lines = await self._build_reserved_command_lines()
         commands_section = (
-            "\n".join(command_lines) if command_lines else "暂无启用的内置指令"
+            "\n".join(command_lines)
+            if command_lines
+            else t(self.context, "help.no_enabled_builtin_commands")
         )
 
         msg_parts = [
             f"AstrBot v{VERSION}(WebUI: {dashboard_version})",
-            "内置指令:",
             commands_section,
         ]
         if notice:

astrbot/builtin_stars/builtin_commands/commands/llm.py

@@ -1,20 +0,0 @@
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageChain
-
-
-class LLMCommands:
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    async def llm(self, event: AstrMessageEvent) -> None:
-        """开启/关闭 LLM"""
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        enable = cfg["provider_settings"].get("enable", True)
-        if enable:
-            cfg["provider_settings"]["enable"] = False
-            status = "关闭"
-        else:
-            cfg["provider_settings"]["enable"] = True
-            status = "开启"
-        cfg.save_config()
-        await event.send(MessageChain().message(f"{status} LLM 聊天功能。"))

astrbot/builtin_stars/builtin_commands/commands/persona.py

@@ -1,216 +0,0 @@
-import builtins
-from typing import TYPE_CHECKING
-
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageEventResult
-
-if TYPE_CHECKING:
-    from astrbot.core.db.po import Persona
-
-
-class PersonaCommands:
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    def _build_tree_output(
-        self,
-        folder_tree: list[dict],
-        all_personas: list["Persona"],
-        depth: int = 0,
-    ) -> list[str]:
-        """递归构建树状输出，使用短线条表示层级"""
-        lines: list[str] = []
-        # 使用短线条作为缩进前缀，每层只用 "│" 加一个空格
-        prefix = "│ " * depth
-
-        for folder in folder_tree:
-            # 输出文件夹
-            lines.append(f"{prefix}├ 📁 {folder['name']}/")
-
-            # 获取该文件夹下的人格
-            folder_personas = [
-                p for p in all_personas if p.folder_id == folder["folder_id"]
-            ]
-            child_prefix = "│ " * (depth + 1)
-
-            # 输出该文件夹下的人格
-            for persona in folder_personas:
-                lines.append(f"{child_prefix}├ 👤 {persona.persona_id}")
-
-            # 递归处理子文件夹
-            children = folder.get("children", [])
-            if children:
-                lines.extend(
-                    self._build_tree_output(
-                        children,
-                        all_personas,
-                        depth + 1,
-                    )
-                )
-
-        return lines
-
-    async def persona(self, message: AstrMessageEvent) -> None:
-        l = message.message_str.split(" ")  # noqa: E741
-        umo = message.unified_msg_origin
-
-        curr_persona_name = "无"
-        cid = await self.context.conversation_manager.get_curr_conversation_id(umo)
-        default_persona = await self.context.persona_manager.get_default_persona_v3(
-            umo=umo,
-        )
-        force_applied_persona_id = None
-
-        curr_cid_title = "无"
-        if cid:
-            conv = await self.context.conversation_manager.get_conversation(
-                unified_msg_origin=umo,
-                conversation_id=cid,
-                create_if_not_exists=True,
-            )
-            if conv is None:
-                message.set_result(
-                    MessageEventResult().message(
-                        "当前对话不存在，请先使用 /new 新建一个对话。",
-                    ),
-                )
-                return
-
-            provider_settings = self.context.get_config(umo=umo).get(
-                "provider_settings",
-                {},
-            )
-            (
-                persona_id,
-                _,
-                force_applied_persona_id,
-                _,
-            ) = await self.context.persona_manager.resolve_selected_persona(
-                umo=umo,
-                conversation_persona_id=conv.persona_id,
-                platform_name=message.get_platform_name(),
-                provider_settings=provider_settings,
-            )
-
-            if persona_id == "[%None]":
-                curr_persona_name = "无"
-            elif persona_id:
-                curr_persona_name = persona_id
-
-            if force_applied_persona_id:
-                curr_persona_name = f"{curr_persona_name} (自定义规则)"
-
-            curr_cid_title = conv.title if conv.title else "新对话"
-            curr_cid_title += f"({cid[:4]})"
-
-        if len(l) == 1:
-            message.set_result(
-                MessageEventResult()
-                .message(
-                    f"""[Persona]
-
-- 人格情景列表: `/persona list`
-- 设置人格情景: `/persona 人格`
-- 人格情景详细信息: `/persona view 人格`
-- 取消人格: `/persona unset`
-
-默认人格情景: {default_persona["name"]}
-当前对话 {curr_cid_title} 的人格情景: {curr_persona_name}
-
-配置人格情景请前往管理面板-配置页
-""",
-                )
-                .use_t2i(False),
-            )
-        elif l[1] == "list":
-            # 获取文件夹树和所有人格
-            folder_tree = await self.context.persona_manager.get_folder_tree()
-            all_personas = self.context.persona_manager.personas
-
-            lines = ["📂 人格列表：\n"]
-
-            # 构建树状输出
-            tree_lines = self._build_tree_output(folder_tree, all_personas)
-            lines.extend(tree_lines)
-
-            # 输出根目录下的人格（没有文件夹的）
-            root_personas = [p for p in all_personas if p.folder_id is None]
-            if root_personas:
-                if tree_lines:  # 如果有文件夹内容，加个空行
-                    lines.append("")
-                for persona in root_personas:
-                    lines.append(f"👤 {persona.persona_id}")
-
-            # 统计信息
-            total_count = len(all_personas)
-            lines.append(f"\n共 {total_count} 个人格")
-            lines.append("\n*使用 `/persona <人格名>` 设置人格")
-            lines.append("*使用 `/persona view <人格名>` 查看详细信息")
-
-            msg = "\n".join(lines)
-            message.set_result(MessageEventResult().message(msg).use_t2i(False))
-        elif l[1] == "view":
-            if len(l) == 2:
-                message.set_result(MessageEventResult().message("请输入人格情景名"))
-                return
-            ps = l[2].strip()
-            if persona := next(
-                builtins.filter(
-                    lambda persona: persona["name"] == ps,
-                    self.context.provider_manager.personas,
-                ),
-                None,
-            ):
-                msg = f"人格{ps}的详细信息：\n"
-                msg += f"{persona['prompt']}\n"
-            else:
-                msg = f"人格{ps}不存在"
-            message.set_result(MessageEventResult().message(msg))
-        elif l[1] == "unset":
-            if not cid:
-                message.set_result(
-                    MessageEventResult().message("当前没有对话，无法取消人格。"),
-                )
-                return
-            await self.context.conversation_manager.update_conversation_persona_id(
-                message.unified_msg_origin,
-                "[%None]",
-            )
-            message.set_result(MessageEventResult().message("取消人格成功。"))
-        else:
-            ps = "".join(l[1:]).strip()
-            if not cid:
-                message.set_result(
-                    MessageEventResult().message(
-                        "当前没有对话，请先开始对话或使用 /new 创建一个对话。",
-                    ),
-                )
-                return
-            if persona := next(
-                builtins.filter(
-                    lambda persona: persona["name"] == ps,
-                    self.context.provider_manager.personas,
-                ),
-                None,
-            ):
-                await self.context.conversation_manager.update_conversation_persona_id(
-                    message.unified_msg_origin,
-                    ps,
-                )
-                force_warn_msg = ""
-                if force_applied_persona_id:
-                    force_warn_msg = (
-                        "提醒：由于自定义规则，您现在切换的人格将不会生效。"
-                    )
-
-                message.set_result(
-                    MessageEventResult().message(
-                        f"设置成功。如果您正在切换到不同的人格，请注意使用 /reset 来清空上下文，防止原人格对话影响现人格。{force_warn_msg}",
-                    ),
-                )
-            else:
-                message.set_result(
-                    MessageEventResult().message(
-                        "不存在该人格情景。使用 /persona list 查看所有。",
-                    ),
-                )

astrbot/builtin_stars/builtin_commands/commands/plugin.py

@@ -1,120 +0,0 @@
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageEventResult
-from astrbot.core import DEMO_MODE, logger
-from astrbot.core.star.filter.command import CommandFilter
-from astrbot.core.star.filter.command_group import CommandGroupFilter
-from astrbot.core.star.star_handler import StarHandlerMetadata, star_handlers_registry
-from astrbot.core.star.star_manager import PluginManager
-
-
-class PluginCommands:
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    async def plugin_ls(self, event: AstrMessageEvent) -> None:
-        """获取已经安装的插件列表。"""
-        parts = ["已加载的插件：\n"]
-        for plugin in self.context.get_all_stars():
-            line = f"- `{plugin.name}` By {plugin.author}: {plugin.desc}"
-            if not plugin.activated:
-                line += " (未启用)"
-            parts.append(line + "\n")
-
-        if len(parts) == 1:
-            plugin_list_info = "没有加载任何插件。"
-        else:
-            plugin_list_info = "".join(parts)
-
-        plugin_list_info += "\n使用 /plugin help <插件名> 查看插件帮助和加载的指令。\n使用 /plugin on/off <插件名> 启用或者禁用插件。"
-        event.set_result(
-            MessageEventResult().message(f"{plugin_list_info}").use_t2i(False),
-        )
-
-    async def plugin_off(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """禁用插件"""
-        if DEMO_MODE:
-            event.set_result(MessageEventResult().message("演示模式下无法禁用插件。"))
-            return
-        if not plugin_name:
-            event.set_result(
-                MessageEventResult().message("/plugin off <插件名> 禁用插件。"),
-            )
-            return
-        await self.context._star_manager.turn_off_plugin(plugin_name)  # type: ignore
-        event.set_result(MessageEventResult().message(f"插件 {plugin_name} 已禁用。"))
-
-    async def plugin_on(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """启用插件"""
-        if DEMO_MODE:
-            event.set_result(MessageEventResult().message("演示模式下无法启用插件。"))
-            return
-        if not plugin_name:
-            event.set_result(
-                MessageEventResult().message("/plugin on <插件名> 启用插件。"),
-            )
-            return
-        await self.context._star_manager.turn_on_plugin(plugin_name)  # type: ignore
-        event.set_result(MessageEventResult().message(f"插件 {plugin_name} 已启用。"))
-
-    async def plugin_get(self, event: AstrMessageEvent, plugin_repo: str = "") -> None:
-        """安装插件"""
-        if DEMO_MODE:
-            event.set_result(MessageEventResult().message("演示模式下无法安装插件。"))
-            return
-        if not plugin_repo:
-            event.set_result(
-                MessageEventResult().message("/plugin get <插件仓库地址> 安装插件"),
-            )
-            return
-        logger.info(f"准备从 {plugin_repo} 安装插件。")
-        if self.context._star_manager:
-            star_mgr: PluginManager = self.context._star_manager
-            try:
-                await star_mgr.install_plugin(plugin_repo)  # type: ignore
-                event.set_result(MessageEventResult().message("安装插件成功。"))
-            except Exception as e:
-                logger.error(f"安装插件失败: {e}")
-                event.set_result(MessageEventResult().message(f"安装插件失败: {e}"))
-                return
-
-    async def plugin_help(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """获取插件帮助"""
-        if not plugin_name:
-            event.set_result(
-                MessageEventResult().message("/plugin help <插件名> 查看插件信息。"),
-            )
-            return
-        plugin = self.context.get_registered_star(plugin_name)
-        if plugin is None:
-            event.set_result(MessageEventResult().message("未找到此插件。"))
-            return
-        help_msg = ""
-        help_msg += f"\n\n✨ 作者: {plugin.author}\n✨ 版本: {plugin.version}"
-        command_handlers = []
-        command_names = []
-        for handler in star_handlers_registry:
-            assert isinstance(handler, StarHandlerMetadata)
-            if handler.handler_module_path != plugin.module_path:
-                continue
-            for filter_ in handler.event_filters:
-                if isinstance(filter_, CommandFilter):
-                    command_handlers.append(handler)
-                    command_names.append(filter_.command_name)
-                    break
-                if isinstance(filter_, CommandGroupFilter):
-                    command_handlers.append(handler)
-                    command_names.append(filter_.group_name)
-
-        if len(command_handlers) > 0:
-            parts = ["\n\n🔧 指令列表：\n"]
-            for i in range(len(command_handlers)):
-                line = f"- {command_names[i]}"
-                if command_handlers[i].desc:
-                    line += f": {command_handlers[i].desc}"
-                parts.append(line + "\n")
-            parts.append("\nTip: 指令的触发需要添加唤醒前缀，默认为 /。")
-            help_msg += "".join(parts)
-
-        ret = f"🧩 插件 {plugin_name} 帮助信息：\n" + help_msg
-        ret += "更多帮助信息请查看插件仓库 README。"
-        event.set_result(MessageEventResult().message(ret).use_t2i(False))

astrbot/builtin_stars/builtin_commands/commands/provider.py

@@ -1,736 +0,0 @@
-from __future__ import annotations
-
-import asyncio
-import time
-from collections.abc import Sequence
-from dataclasses import dataclass
-from typing import TYPE_CHECKING
-
-from astrbot import logger
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageEventResult
-from astrbot.core.provider.entities import ProviderType
-from astrbot.core.utils.error_redaction import safe_error
-
-if TYPE_CHECKING:
-    from astrbot.core.provider.provider import Provider
-
-
-MODEL_LIST_CACHE_TTL_SECONDS_DEFAULT = 30.0
-MODEL_LOOKUP_MAX_CONCURRENCY_DEFAULT = 4
-MODEL_LOOKUP_MAX_CONCURRENCY_UPPER_BOUND = 16
-MODEL_LIST_CACHE_TTL_KEY = "model_list_cache_ttl_seconds"
-MODEL_LOOKUP_MAX_CONCURRENCY_KEY = "model_lookup_max_concurrency"
-MODEL_CACHE_MAX_ENTRIES = 512
-
-
-@dataclass(frozen=True)
-class _ModelLookupConfig:
-    umo: str | None
-    cache_ttl_seconds: float
-    max_concurrency: int
-
-
-class _ModelCache:
-    def __init__(self) -> None:
-        self._store: dict[tuple[str, str | None], tuple[float, list[str]]] = {}
-
-    def get(self, provider_id: str, umo: str | None, ttl: float) -> list[str] | None:
-        if ttl <= 0:
-            return None
-        entry = self._store.get((provider_id, umo))
-        if not entry:
-            return None
-        timestamp, models = entry
-        if time.monotonic() - timestamp > ttl:
-            self._store.pop((provider_id, umo), None)
-            return None
-        return models
-
-    def set(
-        self, provider_id: str, umo: str | None, models: list[str], ttl: float
-    ) -> None:
-        if ttl <= 0:
-            return
-        self._store[(provider_id, umo)] = (time.monotonic(), list(models))
-        self._evict_if_needed()
-
-    def _evict_if_needed(self) -> None:
-        if len(self._store) <= MODEL_CACHE_MAX_ENTRIES:
-            return
-        # Drop oldest entries first when cache grows too large.
-        overflow = len(self._store) - MODEL_CACHE_MAX_ENTRIES
-        for key, _ in sorted(
-            self._store.items(),
-            key=lambda item: item[1][0],
-        )[:overflow]:
-            self._store.pop(key, None)
-
-    def invalidate(
-        self, provider_id: str | None = None, *, umo: str | None = None
-    ) -> None:
-        if provider_id is None:
-            self._store.clear()
-            return
-        if umo is not None:
-            self._store.pop((provider_id, umo), None)
-            return
-        stale_keys = [
-            cache_key for cache_key in self._store if cache_key[0] == provider_id
-        ]
-        for cache_key in stale_keys:
-            self._store.pop(cache_key, None)
-
-
-class ProviderCommands:
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-        self._model_cache = _ModelCache()
-        self._register_provider_change_hook()
-
-    def _register_provider_change_hook(self) -> None:
-        set_change_callback = getattr(
-            self.context.provider_manager,
-            "set_provider_change_callback",
-            None,
-        )
-        if callable(set_change_callback):
-            set_change_callback(self._on_provider_manager_changed)
-            return
-        register_change_hook = getattr(
-            self.context.provider_manager,
-            "register_provider_change_hook",
-            None,
-        )
-        if callable(register_change_hook):
-            register_change_hook(self._on_provider_manager_changed)
-
-    def invalidate_provider_models_cache(
-        self, provider_id: str | None = None, *, umo: str | None = None
-    ) -> None:
-        """Public hook for cache invalidation on external provider config changes."""
-        self._model_cache.invalidate(provider_id, umo=umo)
-
-    def _on_provider_manager_changed(
-        self,
-        provider_id: str,
-        provider_type: ProviderType,
-        umo: str | None,
-    ) -> None:
-        if provider_type == ProviderType.CHAT_COMPLETION:
-            self.invalidate_provider_models_cache(provider_id, umo=umo)
-
-    def _get_provider_settings(self, umo: str | None) -> dict:
-        if not umo:
-            return {}
-        try:
-            return self.context.get_config(umo).get("provider_settings", {}) or {}
-        except Exception as e:
-            logger.debug(
-                "读取 provider_settings 失败，使用默认值: %s",
-                safe_error("", e),
-            )
-            return {}
-
-    def _get_model_cache_ttl(self, umo: str | None) -> float:
-        settings = self._get_provider_settings(umo)
-        raw = settings.get(
-            MODEL_LIST_CACHE_TTL_KEY,
-            MODEL_LIST_CACHE_TTL_SECONDS_DEFAULT,
-        )
-        try:
-            return max(float(raw), 0.0)
-        except Exception as e:
-            logger.debug(
-                "读取 %s 失败，回退默认值 %r: %s",
-                MODEL_LIST_CACHE_TTL_KEY,
-                MODEL_LIST_CACHE_TTL_SECONDS_DEFAULT,
-                safe_error("", e),
-            )
-            return MODEL_LIST_CACHE_TTL_SECONDS_DEFAULT
-
-    def _get_model_lookup_concurrency(self, umo: str | None) -> int:
-        settings = self._get_provider_settings(umo)
-        raw = settings.get(
-            MODEL_LOOKUP_MAX_CONCURRENCY_KEY,
-            MODEL_LOOKUP_MAX_CONCURRENCY_DEFAULT,
-        )
-        try:
-            value = int(raw)
-        except Exception as e:
-            logger.debug(
-                "读取 %s 失败，回退默认值 %r: %s",
-                MODEL_LOOKUP_MAX_CONCURRENCY_KEY,
-                MODEL_LOOKUP_MAX_CONCURRENCY_DEFAULT,
-                safe_error("", e),
-            )
-            value = MODEL_LOOKUP_MAX_CONCURRENCY_DEFAULT
-        return min(max(value, 1), MODEL_LOOKUP_MAX_CONCURRENCY_UPPER_BOUND)
-
-    def _get_model_lookup_config(self, umo: str | None) -> _ModelLookupConfig:
-        return _ModelLookupConfig(
-            umo=umo,
-            cache_ttl_seconds=self._get_model_cache_ttl(umo),
-            max_concurrency=self._get_model_lookup_concurrency(umo),
-        )
-
-    def _resolve_model_name(
-        self,
-        model_name: str,
-        models: Sequence[str],
-    ) -> str | None:
-        """Resolve model name with precedence:
-        exact > case-insensitive > provider-qualified suffix.
-        """
-        requested = model_name.strip()
-        if not requested:
-            return None
-
-        requested_norm = requested.casefold()
-
-        # exact / case-insensitive match
-        for candidate in models:
-            if candidate == requested or candidate.casefold() == requested_norm:
-                return candidate
-
-        # provider-qualified suffix match:
-        # e.g. candidate `openai/gpt-4o` should match requested `gpt-4o`.
-        for candidate in models:
-            cand_norm = candidate.casefold()
-            if cand_norm.endswith(f"/{requested_norm}") or cand_norm.endswith(
-                f":{requested_norm}"
-            ):
-                return candidate
-
-        return None
-
-    def _apply_model(
-        self, prov: Provider, model_name: str, *, umo: str | None = None
-    ) -> str:
-        prov.set_model(model_name)
-        self.invalidate_provider_models_cache(prov.meta().id, umo=umo)
-        return f"切换模型成功。当前提供商: [{prov.meta().id}] 当前模型: [{prov.get_model()}]"
-
-    async def _get_provider_models(
-        self,
-        provider: Provider,
-        *,
-        config: _ModelLookupConfig,
-        use_cache: bool = True,
-    ) -> list[str]:
-        provider_id = provider.meta().id
-        ttl_seconds = config.cache_ttl_seconds
-        umo = config.umo
-        if use_cache:
-            cached = self._model_cache.get(provider_id, umo, ttl_seconds)
-            if cached is not None:
-                return cached
-
-        models = list(await provider.get_models())
-        if use_cache:
-            self._model_cache.set(provider_id, umo, models, ttl_seconds)
-        return models
-
-    async def _get_models_or_reply_error(
-        self,
-        message: AstrMessageEvent,
-        prov: Provider,
-        config: _ModelLookupConfig,
-        *,
-        error_prefix: str,
-        disable_t2i: bool = False,
-        warning_log: str | None = None,
-    ) -> list[str] | None:
-        try:
-            return await self._get_provider_models(prov, config=config)
-        except asyncio.CancelledError:
-            raise
-        except Exception as e:
-            if warning_log is not None:
-                logger.warning(
-                    warning_log,
-                    prov.meta().id,
-                    safe_error("", e),
-                )
-            result = MessageEventResult().message(safe_error(error_prefix, e))
-            if disable_t2i:
-                result = result.use_t2i(False)
-            message.set_result(result)
-            return None
-
-    def _log_reachability_failure(
-        self,
-        provider,
-        provider_capability_type: ProviderType | None,
-        err_code: str,
-        err_reason: str,
-    ) -> None:
-        """记录不可达原因到日志。"""
-        meta = provider.meta()
-        logger.warning(
-            "Provider reachability check failed: id=%s type=%s code=%s reason=%s",
-            meta.id,
-            provider_capability_type.name if provider_capability_type else "unknown",
-            err_code,
-            err_reason,
-        )
-
-    async def _test_provider_capability(self, provider):
-        """测试单个 provider 的可用性"""
-        meta = provider.meta()
-        provider_capability_type = meta.provider_type
-
-        try:
-            await provider.test()
-            return True, None, None
-        except Exception as e:
-            err_code = "TEST_FAILED"
-            err_reason = safe_error("", e)
-            self._log_reachability_failure(
-                provider, provider_capability_type, err_code, err_reason
-            )
-            return False, err_code, err_reason
-
-    async def _find_provider_for_model(
-        self,
-        model_name: str,
-        *,
-        exclude_provider_id: str | None = None,
-        config: _ModelLookupConfig,
-        use_cache: bool = True,
-    ) -> tuple[Provider | None, str | None]:
-        all_providers = []
-        for provider in self.context.get_all_providers():
-            provider_meta = provider.meta()
-            if provider_meta.provider_type != ProviderType.CHAT_COMPLETION:
-                continue
-            if (
-                exclude_provider_id is not None
-                and provider_meta.id == exclude_provider_id
-            ):
-                continue
-            all_providers.append(provider)
-        if not all_providers:
-            return None, None
-
-        semaphore = asyncio.Semaphore(config.max_concurrency)
-
-        async def fetch_models(
-            provider: Provider,
-        ) -> tuple[Provider, list[str] | None, str | None]:
-            async with semaphore:
-                try:
-                    models = await self._get_provider_models(
-                        provider,
-                        config=config,
-                        use_cache=use_cache,
-                    )
-                    return provider, models, None
-                except asyncio.CancelledError:
-                    raise
-                except Exception as e:
-                    err = safe_error("", e)
-                    logger.debug(
-                        "跨提供商查找模型 %s 获取 %s 模型列表失败: %s",
-                        model_name,
-                        provider.meta().id,
-                        err,
-                    )
-                    return provider, None, err
-
-        results = await asyncio.gather(
-            *(fetch_models(provider) for provider in all_providers)
-        )
-        failed_provider_errors: list[tuple[str, str]] = []
-        for provider, models, err in results:
-            if err is not None:
-                failed_provider_errors.append((provider.meta().id, err))
-                continue
-            if models is None:
-                continue
-
-            matched_model_name = self._resolve_model_name(model_name, models)
-            if matched_model_name is not None:
-                return provider, matched_model_name
-
-        if failed_provider_errors and len(failed_provider_errors) == len(all_providers):
-            failed_ids = ",".join(
-                provider_id for provider_id, _ in failed_provider_errors
-            )
-            logger.error(
-                "跨提供商查找模型 %s 时，所有 %d 个提供商的 get_models() 均失败: %s。请检查配置或网络",
-                model_name,
-                len(all_providers),
-                failed_ids,
-            )
-        elif failed_provider_errors:
-            logger.debug(
-                "跨提供商查找模型 %s 时有 %d 个提供商获取模型失败: %s",
-                model_name,
-                len(failed_provider_errors),
-                ",".join(
-                    f"{provider_id}({error})"
-                    for provider_id, error in failed_provider_errors
-                ),
-            )
-        return None, None
-
-    async def provider(
-        self,
-        event: AstrMessageEvent,
-        idx: str | int | None = None,
-        idx2: int | None = None,
-    ) -> None:
-        """查看或者切换 LLM Provider"""
-        umo = event.unified_msg_origin
-        cfg = self.context.get_config(umo).get("provider_settings", {})
-        reachability_check_enabled = cfg.get("reachability_check", True)
-
-        if idx is None:
-            parts = ["## 载入的 LLM 提供商\n"]
-
-            # 获取所有类型的提供商
-            llms = list(self.context.get_all_providers())
-            ttss = self.context.get_all_tts_providers()
-            stts = self.context.get_all_stt_providers()
-
-            # 构造待检测列表: [(provider, type_label), ...]
-            all_providers = []
-            all_providers.extend([(p, "llm") for p in llms])
-            all_providers.extend([(p, "tts") for p in ttss])
-            all_providers.extend([(p, "stt") for p in stts])
-
-            # 并发测试连通性
-            if reachability_check_enabled:
-                if all_providers:
-                    await event.send(
-                        MessageEventResult().message(
-                            "正在进行提供商可达性测试，请稍候..."
-                        )
-                    )
-                check_results = await asyncio.gather(
-                    *[self._test_provider_capability(p) for p, _ in all_providers],
-                    return_exceptions=True,
-                )
-            else:
-                # 用 None 表示未检测
-                check_results = [None for _ in all_providers]
-
-            # 整合结果
-            display_data = []
-            for (p, p_type), reachable in zip(all_providers, check_results):
-                meta = p.meta()
-                id_ = meta.id
-                error_code = None
-
-                if isinstance(reachable, asyncio.CancelledError):
-                    raise reachable
-                if isinstance(reachable, Exception):
-                    # 异常情况下兜底处理，避免单个 provider 导致列表失败
-                    self._log_reachability_failure(
-                        p,
-                        None,
-                        reachable.__class__.__name__,
-                        safe_error("", reachable),
-                    )
-                    reachable_flag = False
-                    error_code = reachable.__class__.__name__
-                elif isinstance(reachable, tuple):
-                    reachable_flag, error_code, _ = reachable
-                else:
-                    reachable_flag = reachable
-
-                # 根据类型构建显示名称
-                if p_type == "llm":
-                    info = f"{id_} ({meta.model})"
-                else:
-                    info = f"{id_}"
-
-                # 确定状态标记
-                if reachable_flag is True:
-                    mark = " ✅"
-                elif reachable_flag is False:
-                    if error_code:
-                        mark = f" ❌(错误码: {error_code})"
-                    else:
-                        mark = " ❌"
-                else:
-                    mark = ""  # 不支持检测时不显示标记
-
-                display_data.append(
-                    {
-                        "type": p_type,
-                        "info": info,
-                        "mark": mark,
-                        "provider": p,
-                    }
-                )
-
-            # 分组输出
-            # 1. LLM
-            llm_data = [d for d in display_data if d["type"] == "llm"]
-            for i, d in enumerate(llm_data):
-                line = f"{i + 1}. {d['info']}{d['mark']}"
-                provider_using = self.context.get_using_provider(umo=umo)
-                if (
-                    provider_using
-                    and provider_using.meta().id == d["provider"].meta().id
-                ):
-                    line += " (当前使用)"
-                parts.append(line + "\n")
-
-            # 2. TTS
-            tts_data = [d for d in display_data if d["type"] == "tts"]
-            if tts_data:
-                parts.append("\n## 载入的 TTS 提供商\n")
-                for i, d in enumerate(tts_data):
-                    line = f"{i + 1}. {d['info']}{d['mark']}"
-                    tts_using = self.context.get_using_tts_provider(umo=umo)
-                    if tts_using and tts_using.meta().id == d["provider"].meta().id:
-                        line += " (当前使用)"
-                    parts.append(line + "\n")
-
-            # 3. STT
-            stt_data = [d for d in display_data if d["type"] == "stt"]
-            if stt_data:
-                parts.append("\n## 载入的 STT 提供商\n")
-                for i, d in enumerate(stt_data):
-                    line = f"{i + 1}. {d['info']}{d['mark']}"
-                    stt_using = self.context.get_using_stt_provider(umo=umo)
-                    if stt_using and stt_using.meta().id == d["provider"].meta().id:
-                        line += " (当前使用)"
-                    parts.append(line + "\n")
-
-            parts.append("\n使用 /provider <序号> 切换 LLM 提供商。")
-            ret = "".join(parts)
-
-            if ttss:
-                ret += "\n使用 /provider tts <序号> 切换 TTS 提供商。"
-            if stts:
-                ret += "\n使用 /provider stt <序号> 切换 STT 提供商。"
-            if not reachability_check_enabled:
-                ret += "\n已跳过提供商可达性检测，如需检测请在配置文件中开启。"
-
-            event.set_result(MessageEventResult().message(ret))
-        elif idx == "tts":
-            if idx2 is None:
-                event.set_result(MessageEventResult().message("请输入序号。"))
-                return
-            if idx2 > len(self.context.get_all_tts_providers()) or idx2 < 1:
-                event.set_result(MessageEventResult().message("无效的提供商序号。"))
-                return
-            provider = self.context.get_all_tts_providers()[idx2 - 1]
-            id_ = provider.meta().id
-            await self.context.provider_manager.set_provider(
-                provider_id=id_,
-                provider_type=ProviderType.TEXT_TO_SPEECH,
-                umo=umo,
-            )
-            event.set_result(MessageEventResult().message(f"成功切换到 {id_}。"))
-        elif idx == "stt":
-            if idx2 is None:
-                event.set_result(MessageEventResult().message("请输入序号。"))
-                return
-            if idx2 > len(self.context.get_all_stt_providers()) or idx2 < 1:
-                event.set_result(MessageEventResult().message("无效的提供商序号。"))
-                return
-            provider = self.context.get_all_stt_providers()[idx2 - 1]
-            id_ = provider.meta().id
-            await self.context.provider_manager.set_provider(
-                provider_id=id_,
-                provider_type=ProviderType.SPEECH_TO_TEXT,
-                umo=umo,
-            )
-            event.set_result(MessageEventResult().message(f"成功切换到 {id_}。"))
-        elif isinstance(idx, int):
-            if idx > len(self.context.get_all_providers()) or idx < 1:
-                event.set_result(MessageEventResult().message("无效的提供商序号。"))
-                return
-            provider = self.context.get_all_providers()[idx - 1]
-            id_ = provider.meta().id
-            await self.context.provider_manager.set_provider(
-                provider_id=id_,
-                provider_type=ProviderType.CHAT_COMPLETION,
-                umo=umo,
-            )
-            event.set_result(MessageEventResult().message(f"成功切换到 {id_}。"))
-        else:
-            event.set_result(MessageEventResult().message("无效的参数。"))
-
-    async def _switch_model_by_name(
-        self, message: AstrMessageEvent, model_name: str, prov: Provider
-    ) -> None:
-        model_name = model_name.strip()
-        if not model_name:
-            message.set_result(MessageEventResult().message("模型名不能为空。"))
-            return
-
-        umo = message.unified_msg_origin
-        config = self._get_model_lookup_config(umo)
-        curr_provider_id = prov.meta().id
-
-        models = await self._get_models_or_reply_error(
-            message,
-            prov,
-            config,
-            error_prefix="获取当前提供商模型列表失败: ",
-            warning_log="获取当前提供商 %s 模型列表失败，停止跨提供商查找: %s",
-        )
-        if models is None:
-            return
-
-        matched_model_name = self._resolve_model_name(model_name, models)
-        if matched_model_name is not None:
-            message.set_result(
-                MessageEventResult().message(
-                    self._apply_model(prov, matched_model_name, umo=umo)
-                ),
-            )
-            return
-
-        target_prov, matched_target_model_name = await self._find_provider_for_model(
-            model_name,
-            exclude_provider_id=curr_provider_id,
-            config=config,
-        )
-
-        if target_prov is None or matched_target_model_name is None:
-            message.set_result(
-                MessageEventResult().message(
-                    f"模型 [{model_name}] 未在任何已配置的提供商中找到，或所有提供商模型列表获取失败，请检查配置或网络后重试。",
-                ),
-            )
-            return
-
-        target_id = target_prov.meta().id
-        try:
-            await self.context.provider_manager.set_provider(
-                provider_id=target_id,
-                provider_type=ProviderType.CHAT_COMPLETION,
-                umo=umo,
-            )
-            self._apply_model(target_prov, matched_target_model_name, umo=umo)
-            message.set_result(
-                MessageEventResult().message(
-                    f"检测到模型 [{matched_target_model_name}] 属于提供商 [{target_id}]，已自动切换提供商并设置模型。",
-                ),
-            )
-        except asyncio.CancelledError:
-            raise
-        except Exception as e:
-            message.set_result(
-                MessageEventResult().message(
-                    safe_error("跨提供商切换并设置模型失败: ", e)
-                ),
-            )
-
-    async def model_ls(
-        self,
-        message: AstrMessageEvent,
-        idx_or_name: int | str | None = None,
-    ) -> None:
-        """查看或者切换模型"""
-        prov = self.context.get_using_provider(message.unified_msg_origin)
-        if not prov:
-            message.set_result(
-                MessageEventResult().message("未找到任何 LLM 提供商。请先配置。"),
-            )
-            return
-        config = self._get_model_lookup_config(message.unified_msg_origin)
-
-        if idx_or_name is None:
-            models = await self._get_models_or_reply_error(
-                message,
-                prov,
-                config,
-                error_prefix="获取模型列表失败: ",
-                disable_t2i=True,
-            )
-            if models is None:
-                return
-            parts = ["下面列出了此模型提供商可用模型:"]
-            for i, model in enumerate(models, 1):
-                parts.append(f"\n{i}. {model}")
-
-            curr_model = prov.get_model() or "无"
-            parts.append(f"\n当前模型: [{curr_model}]")
-            parts.append(
-                "\nTips: 使用 /model <模型名/编号> 切换模型。输入模型名时可自动跨提供商查找并切换；跨提供商也可使用 /provider 切换。"
-            )
-
-            ret = "".join(parts)
-            message.set_result(MessageEventResult().message(ret).use_t2i(False))
-        elif isinstance(idx_or_name, int):
-            models = await self._get_models_or_reply_error(
-                message,
-                prov,
-                config,
-                error_prefix="获取模型列表失败: ",
-            )
-            if models is None:
-                return
-            if idx_or_name > len(models) or idx_or_name < 1:
-                message.set_result(MessageEventResult().message("模型序号错误。"))
-            else:
-                try:
-                    new_model = models[idx_or_name - 1]
-                    message.set_result(
-                        MessageEventResult().message(
-                            self._apply_model(
-                                prov,
-                                new_model,
-                                umo=message.unified_msg_origin,
-                            )
-                        ),
-                    )
-                except Exception as e:
-                    message.set_result(
-                        MessageEventResult().message(
-                            safe_error("切换模型未知错误: ", e)
-                        ),
-                    )
-                    return
-        else:
-            await self._switch_model_by_name(message, idx_or_name, prov)
-
-    async def key(self, message: AstrMessageEvent, index: int | None = None) -> None:
-        prov = self.context.get_using_provider(message.unified_msg_origin)
-        if not prov:
-            message.set_result(
-                MessageEventResult().message("未找到任何 LLM 提供商。请先配置。"),
-            )
-            return
-
-        if index is None:
-            keys_data = prov.get_keys()
-            curr_key = prov.get_current_key()
-            parts = ["Key:"]
-            for i, k in enumerate(keys_data, 1):
-                parts.append(f"\n{i}. {k[:8]}")
-
-            parts.append(f"\n当前 Key: {curr_key[:8]}")
-            parts.append("\n当前模型: " + prov.get_model())
-            parts.append("\n使用 /key <idx> 切换 Key。")
-
-            ret = "".join(parts)
-            message.set_result(MessageEventResult().message(ret).use_t2i(False))
-        else:
-            keys_data = prov.get_keys()
-            if index > len(keys_data) or index < 1:
-                message.set_result(MessageEventResult().message("Key 序号错误。"))
-            else:
-                try:
-                    new_key = keys_data[index - 1]
-                    prov.set_key(new_key)
-                    self.invalidate_provider_models_cache(
-                        prov.meta().id,
-                        umo=message.unified_msg_origin,
-                    )
-                    message.set_result(MessageEventResult().message("切换 Key 成功。"))
-                except Exception as e:
-                    message.set_result(
-                        MessageEventResult().message(
-                            safe_error("切换 Key 未知错误: ", e)
-                        ),
-                    )
-                    return

astrbot/builtin_stars/builtin_commands/commands/setunset.py

@@ -1,6 +1,8 @@
 from astrbot.api import sp, star
 from astrbot.api.event import AstrMessageEvent, MessageEventResult
 
+from ..i18n import t
+
 
 class SetUnsetCommands:
     def __init__(self, context: star.Context) -> None:
@@ -15,7 +17,12 @@ class SetUnsetCommands:
 
         event.set_result(
             MessageEventResult().message(
-                f"会话 {uid} 变量 {key} 存储成功。使用 /unset 移除。",
+                t(
+                    self.context,
+                    "setunset.set_success",
+                    session_id=uid,
+                    key=key,
+                ),
             ),
         )
 
@@ -26,11 +33,20 @@ class SetUnsetCommands:
 
         if key not in session_var:
             event.set_result(
-                MessageEventResult().message("没有那个变量名。格式 /unset 变量名。"),
+                MessageEventResult().message(
+                    t(self.context, "setunset.unset_not_found")
+                ),
             )
         else:
             del session_var[key]
             await sp.session_put(uid, "session_variables", session_var)
             event.set_result(
-                MessageEventResult().message(f"会话 {uid} 变量 {key} 移除成功。"),
+                MessageEventResult().message(
+                    t(
+                        self.context,
+                        "setunset.unset_success",
+                        session_id=uid,
+                        key=key,
+                    ),
+                ),
             )

astrbot/builtin_stars/builtin_commands/commands/sid.py

@@ -3,6 +3,8 @@
 from astrbot.api import star
 from astrbot.api.event import AstrMessageEvent, MessageEventResult
 
+from ..i18n import t
+
 
 class SIDCommand:
     """会话ID命令类"""
@@ -17,20 +19,24 @@ class SIDCommand:
         umo_platform = event.session.platform_id
         umo_msg_type = event.session.message_type.value
         umo_session_id = event.session.session_id
-        ret = (
-            f"UMO: 「{sid}」 此值可用于设置白名单。\n"
-            f"UID: 「{user_id}」 此值可用于设置管理员。\n"
-            f"消息会话来源信息:\n"
-            f"  机器人 ID: 「{umo_platform}」\n"
-            f"  消息类型: 「{umo_msg_type}」\n"
-            f"  会话 ID: 「{umo_session_id}」\n"
-            f"消息来源可用于配置机器人的配置文件路由。"
+        ret = t(
+            self.context,
+            "sid.info",
+            sid=sid,
+            user_id=user_id,
+            platform=umo_platform,
+            message_type=umo_msg_type,
+            session_id=umo_session_id,
         )
 
         if (
             self.context.get_config()["platform_settings"]["unique_session"]
             and event.get_group_id()
         ):
-            ret += f"\n\n当前处于独立会话模式, 此群 ID: 「{event.get_group_id()}」, 也可将此 ID 加入白名单来放行整个群聊。"
+            ret += t(
+                self.context,
+                "sid.group_whitelist",
+                group_id=event.get_group_id(),
+            )
 
         event.set_result(MessageEventResult().message(ret).use_t2i(False))

astrbot/builtin_stars/builtin_commands/commands/t2i.py

@@ -1,23 +0,0 @@
-"""文本转图片命令"""
-
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageEventResult
-
-
-class T2ICommand:
-    """文本转图片命令类"""
-
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    async def t2i(self, event: AstrMessageEvent) -> None:
-        """开关文本转图片"""
-        config = self.context.get_config(umo=event.unified_msg_origin)
-        if config["t2i"]:
-            config["t2i"] = False
-            config.save_config()
-            event.set_result(MessageEventResult().message("已关闭文本转图片模式。"))
-            return
-        config["t2i"] = True
-        config.save_config()
-        event.set_result(MessageEventResult().message("已开启文本转图片模式。"))

astrbot/builtin_stars/builtin_commands/commands/tts.py

@@ -1,36 +0,0 @@
-"""文本转语音命令"""
-
-from astrbot.api import star
-from astrbot.api.event import AstrMessageEvent, MessageEventResult
-from astrbot.core.star.session_llm_manager import SessionServiceManager
-
-
-class TTSCommand:
-    """文本转语音命令类"""
-
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-
-    async def tts(self, event: AstrMessageEvent) -> None:
-        """开关文本转语音（会话级别）"""
-        umo = event.unified_msg_origin
-        ses_tts = await SessionServiceManager.is_tts_enabled_for_session(umo)
-        cfg = self.context.get_config(umo=umo)
-        tts_enable = cfg["provider_tts_settings"]["enable"]
-
-        # 切换状态
-        new_status = not ses_tts
-        await SessionServiceManager.set_tts_status_for_session(umo, new_status)
-
-        status_text = "已开启" if new_status else "已关闭"
-
-        if new_status and not tts_enable:
-            event.set_result(
-                MessageEventResult().message(
-                    f"{status_text}当前会话的文本转语音。但 TTS 功能在配置中未启用，请前往 WebUI 开启。",
-                ),
-            )
-        else:
-            event.set_result(
-                MessageEventResult().message(f"{status_text}当前会话的文本转语音。"),
-            )

astrbot/builtin_stars/builtin_commands/i18n.py

@@ -0,0 +1,30 @@
+import json
+from functools import lru_cache
+from pathlib import Path
+from typing import Any
+
+LOCALE_DIR = Path(__file__).resolve().parent / "locales"
+
+
+@lru_cache(maxsize=2)
+def _load_locale(language: str) -> dict[str, Any]:
+    with (LOCALE_DIR / f"{language}.json").open(encoding="utf-8") as f:
+        return json.load(f)
+
+
+def _resolve_key(data: dict[str, Any], translation_key: str) -> Any:
+    value: Any = data
+    for part in translation_key.split("."):
+        if not isinstance(value, dict) or part not in value:
+            return None
+        value = value[part]
+    return value
+
+
+def t(context: Any, translation_key: str, **kwargs: Any) -> str:
+    text = _resolve_key(_load_locale(context.get_current_language()), translation_key)
+    if not isinstance(text, str):
+        return translation_key
+    if not kwargs:
+        return text
+    return text.format(**kwargs)

astrbot/builtin_stars/builtin_commands/locales/en-US.json

@@ -0,0 +1,33 @@
+{
+  "help": {
+    "no_enabled_builtin_commands": "No enabled built-in commands."
+  },
+  "sid": {
+    "info": "UMO: 「{sid}」\nUID: 「{user_id}」\n*Use UMO to set whitelist and configure routing, use UID to set admin list.\n\nYour session information:\nBot ID: 「{platform}」\nMessage Type: 「{message_type}」\nSession ID: 「{session_id}」\n\n",
+    "group_whitelist": "\n\nThe group's ID: 「{group_id}」. Set this ID to whitelist to allow the entire group."
+  },
+  "dashboard": {
+    "updating": "⏳ Updating dashboard...",
+    "updated": "✅ Dashboard updated successfully."
+  },
+  "scene": {
+    "group_unique_on": "group chat with unique session enabled",
+    "group_unique_off": "group chat with unique session disabled",
+    "private": "private chat"
+  },
+  "conversation": {
+    "reset_admin_required": "Reset command requires admin permission in {scene_name} scenario, you (ID {sender_id}) are not admin, cannot perform this action.",
+    "reset_success": "✅ Conversation reset successfully.",
+    "no_provider": "😕 Cannot find any LLM provider. Configure one first.",
+    "no_conversation": "😕 You are not in a conversation. Use /new to create one.",
+    "stop_requested": "✅ Requested to stop {count} running tasks.",
+    "no_running_tasks": "✅ No running tasks in the current session.",
+    "new_created": "✅ New conversation created.",
+    "switched_new": "✅ Switched to new conversation: {conversation_id}."
+  },
+  "setunset": {
+    "set_success": "Session {session_id} variable {key} saved. Use /unset to remove it.",
+    "unset_not_found": "No variable with that name. Format: /unset variable_name.",
+    "unset_success": "Session {session_id} variable {key} removed."
+  }
+}

astrbot/builtin_stars/builtin_commands/locales/zh-CN.json

@@ -0,0 +1,33 @@
+{
+  "help": {
+    "no_enabled_builtin_commands": "没有已启用的内置指令。"
+  },
+  "sid": {
+    "info": "UMO: 「{sid}」\nUID: 「{user_id}」\n*使用 UMO 设置白名单和配置文件路由，使用 UID 设置管理员列表。\n\n当前会话信息：\n机器人 ID: 「{platform}」\n消息类型: 「{message_type}」\n会话 ID: 「{session_id}」\n\n",
+    "group_whitelist": "\n\n当前群聊 ID: 「{group_id}」。将此 ID 加入白名单可允许整个群聊。"
+  },
+  "dashboard": {
+    "updating": "⏳ 正在更新管理面板...",
+    "updated": "✅ 管理面板更新成功。"
+  },
+  "scene": {
+    "group_unique_on": "群聊+会话隔离开启",
+    "group_unique_off": "群聊+会话隔离关闭",
+    "private": "私聊"
+  },
+  "conversation": {
+    "reset_admin_required": "在 {scene_name} 场景下，reset 指令需要管理员权限。你(ID {sender_id})不是管理员，无法执行此操作。",
+    "reset_success": "✅ 会话已重置。",
+    "no_provider": "😕 未找到可用的 LLM 提供商，请先配置。",
+    "no_conversation": "😕 当前未处于对话状态。请使用 /new 创建一个对话。",
+    "stop_requested": "✅ 已请求停止 {count} 个正在运行的任务。",
+    "no_running_tasks": "✅ 当前会话没有正在运行的任务。",
+    "new_created": "✅ 已创建新对话。",
+    "switched_new": "✅ 已切换到新对话：{conversation_id}。"
+  },
+  "setunset": {
+    "set_success": "会话 {session_id} 变量 {key} 存储成功。使用 /unset 移除。",
+    "unset_not_found": "没有那个变量名。格式 /unset 变量名。",
+    "unset_success": "会话 {session_id} 变量 {key} 移除成功。"
+  }
+}

astrbot/builtin_stars/builtin_commands/main.py

@@ -3,17 +3,10 @@ from astrbot.api.event import AstrMessageEvent, filter
 
 from .commands import (
     AdminCommands,
-    AlterCmdCommands,
     ConversationCommands,
     HelpCommand,
-    LLMCommands,
-    PersonaCommands,
-    PluginCommands,
-    ProviderCommands,
     SetUnsetCommands,
     SIDCommand,
-    T2ICommand,
-    TTSCommand,
 )
 
 
@@ -21,198 +14,49 @@ class Main(star.Star):
     def __init__(self, context: star.Context) -> None:
         self.context = context
 
-        self.help_c = HelpCommand(self.context)
-        self.llm_c = LLMCommands(self.context)
-        self.plugin_c = PluginCommands(self.context)
         self.admin_c = AdminCommands(self.context)
         self.conversation_c = ConversationCommands(self.context)
-        self.provider_c = ProviderCommands(self.context)
-        self.persona_c = PersonaCommands(self.context)
-        self.alter_cmd_c = AlterCmdCommands(self.context)
+        self.help_c = HelpCommand(self.context)
         self.setunset_c = SetUnsetCommands(self.context)
-        self.t2i_c = T2ICommand(self.context)
-        self.tts_c = TTSCommand(self.context)
         self.sid_c = SIDCommand(self.context)
 
     @filter.command("help")
     async def help(self, event: AstrMessageEvent) -> None:
-        """查看帮助"""
+        """Show help message"""
         await self.help_c.help(event)
 
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("llm")
-    async def llm(self, event: AstrMessageEvent) -> None:
-        """开启/关闭 LLM"""
-        await self.llm_c.llm(event)
-
-    @filter.command_group("plugin")
-    def plugin(self) -> None:
-        """插件管理"""
-
-    @plugin.command("ls")
-    async def plugin_ls(self, event: AstrMessageEvent) -> None:
-        """获取已经安装的插件列表。"""
-        await self.plugin_c.plugin_ls(event)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @plugin.command("off")
-    async def plugin_off(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """禁用插件"""
-        await self.plugin_c.plugin_off(event, plugin_name)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @plugin.command("on")
-    async def plugin_on(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """启用插件"""
-        await self.plugin_c.plugin_on(event, plugin_name)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @plugin.command("get")
-    async def plugin_get(self, event: AstrMessageEvent, plugin_repo: str = "") -> None:
-        """安装插件"""
-        await self.plugin_c.plugin_get(event, plugin_repo)
-
-    @plugin.command("help")
-    async def plugin_help(self, event: AstrMessageEvent, plugin_name: str = "") -> None:
-        """获取插件帮助"""
-        await self.plugin_c.plugin_help(event, plugin_name)
-
-    @filter.command("t2i")
-    async def t2i(self, event: AstrMessageEvent) -> None:
-        """开关文本转图片"""
-        await self.t2i_c.t2i(event)
-
-    @filter.command("tts")
-    async def tts(self, event: AstrMessageEvent) -> None:
-        """开关文本转语音（会话级别）"""
-        await self.tts_c.tts(event)
-
     @filter.command("sid")
     async def sid(self, event: AstrMessageEvent) -> None:
-        """获取会话 ID 和 管理员 ID"""
+        """Get session ID and other related information"""
         await self.sid_c.sid(event)
 
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("op")
-    async def op(self, event: AstrMessageEvent, admin_id: str = "") -> None:
-        """授权管理员。op <admin_id>"""
-        await self.admin_c.op(event, admin_id)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("deop")
-    async def deop(self, event: AstrMessageEvent, admin_id: str) -> None:
-        """取消授权管理员。deop <admin_id>"""
-        await self.admin_c.deop(event, admin_id)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("wl")
-    async def wl(self, event: AstrMessageEvent, sid: str = "") -> None:
-        """添加白名单。wl <sid>"""
-        await self.admin_c.wl(event, sid)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("dwl")
-    async def dwl(self, event: AstrMessageEvent, sid: str) -> None:
-        """删除白名单。dwl <sid>"""
-        await self.admin_c.dwl(event, sid)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("provider")
-    async def provider(
-        self,
-        event: AstrMessageEvent,
-        idx: str | int | None = None,
-        idx2: int | None = None,
-    ) -> None:
-        """查看或者切换 LLM Provider"""
-        await self.provider_c.provider(event, idx, idx2)
-
     @filter.command("reset")
     async def reset(self, message: AstrMessageEvent) -> None:
-        """重置 LLM 会话"""
+        """Reset conversation history"""
         await self.conversation_c.reset(message)
 
     @filter.command("stop")
     async def stop(self, message: AstrMessageEvent) -> None:
-        """停止当前会话中正在运行的 Agent"""
+        """Stop agent execution"""
         await self.conversation_c.stop(message)
 
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("model")
-    async def model_ls(
-        self,
-        message: AstrMessageEvent,
-        idx_or_name: int | str | None = None,
-    ) -> None:
-        """查看或者切换模型"""
-        await self.provider_c.model_ls(message, idx_or_name)
-
-    @filter.command("history")
-    async def his(self, message: AstrMessageEvent, page: int = 1) -> None:
-        """查看对话记录"""
-        await self.conversation_c.his(message, page)
-
-    @filter.command("ls")
-    async def convs(self, message: AstrMessageEvent, page: int = 1) -> None:
-        """查看对话列表"""
-        await self.conversation_c.convs(message, page)
-
     @filter.command("new")
     async def new_conv(self, message: AstrMessageEvent) -> None:
-        """创建新对话"""
+        """Create new conversation"""
         await self.conversation_c.new_conv(message)
 
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("groupnew")
-    async def groupnew_conv(self, message: AstrMessageEvent, sid: str) -> None:
-        """创建新群聊对话"""
-        await self.conversation_c.groupnew_conv(message, sid)
-
-    @filter.command("switch")
-    async def switch_conv(
-        self, message: AstrMessageEvent, index: int | None = None
-    ) -> None:
-        """通过 /ls 前面的序号切换对话"""
-        await self.conversation_c.switch_conv(message, index)
-
-    @filter.command("rename")
-    async def rename_conv(self, message: AstrMessageEvent, new_name: str) -> None:
-        """重命名对话"""
-        await self.conversation_c.rename_conv(message, new_name)
-
-    @filter.command("del")
-    async def del_conv(self, message: AstrMessageEvent) -> None:
-        """删除当前对话"""
-        await self.conversation_c.del_conv(message)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("key")
-    async def key(self, message: AstrMessageEvent, index: int | None = None) -> None:
-        """查看或者切换 Key"""
-        await self.provider_c.key(message, index)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("persona")
-    async def persona(self, message: AstrMessageEvent) -> None:
-        """查看或者切换 Persona"""
-        await self.persona_c.persona(message)
-
     @filter.permission_type(filter.PermissionType.ADMIN)
     @filter.command("dashboard_update")
     async def update_dashboard(self, event: AstrMessageEvent) -> None:
-        """更新管理面板"""
+        """Update AstrBot WebUI"""
         await self.admin_c.update_dashboard(event)
 
     @filter.command("set")
     async def set_variable(self, event: AstrMessageEvent, key: str, value: str) -> None:
+        """Set session variable"""
         await self.setunset_c.set_variable(event, key, value)
 
     @filter.command("unset")
     async def unset_variable(self, event: AstrMessageEvent, key: str) -> None:
+        """Unset session variable"""
         await self.setunset_c.unset_variable(event, key)
-
-    @filter.permission_type(filter.PermissionType.ADMIN)
-    @filter.command("alter_cmd", alias={"alter"})
-    async def alter_cmd(self, event: AstrMessageEvent) -> None:
-        """修改命令权限"""
-        await self.alter_cmd_c.alter_cmd(event)

astrbot/builtin_stars/session_controller/main.py

@@ -91,6 +91,8 @@ class Main(Star):
                         controller: SessionController,
                         event: AstrMessageEvent,
                     ) -> None:
+                        if not event.message_str or not event.message_str.strip():
+                            return
                         event.message_obj.message.insert(
                             0,
                             Comp.At(qq=event.get_self_id(), name=event.get_self_id()),

astrbot/builtin_stars/web_searcher/engines/__init__.py

@@ -1,112 +0,0 @@
-import random
-import urllib.parse
-from dataclasses import dataclass
-
-from aiohttp import ClientSession
-from bs4 import BeautifulSoup, Tag
-
-HEADERS = {
-    "User-Agent": "Mozilla/5.0 (Windows NT 6.1; rv:84.0) Gecko/20100101 Firefox/84.0",
-    "Accept": "*/*",
-    "Connection": "keep-alive",
-    "Accept-Language": "en-GB,en;q=0.5",
-}
-
-USER_AGENT_BING = "Mozilla/5.0 (Windows NT 6.1; rv:84.0) Gecko/20100101 Firefox/84.0"
-USER_AGENTS = [
-    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.131 Safari/537.36",
-    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36",
-    "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:89.0) Gecko/20100101 Firefox/89.0",
-    "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:88.0) Gecko/20100101 Firefox/88.0",
-    "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.131 Safari/537.36",
-    "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36",
-    "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Version/14.1.2 Safari/537.36",
-    "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Version/14.1 Safari/537.36",
-    "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:89.0) Gecko/20100101 Firefox/89.0",
-    "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:88.0) Gecko/20100101 Firefox/88.0",
-]
-
-
-@dataclass
-class SearchResult:
-    title: str
-    url: str
-    snippet: str
-    favicon: str | None = None
-
-    def __str__(self) -> str:
-        return f"{self.title} - {self.url}\n{self.snippet}"
-
-
-class SearchEngine:
-    """搜索引擎爬虫基类"""
-
-    def __init__(self) -> None:
-        self.TIMEOUT = 10
-        self.page = 1
-        self.headers = HEADERS
-
-    def _set_selector(self, selector: str) -> str:
-        raise NotImplementedError
-
-    async def _get_next_page(self, query: str) -> str:
-        raise NotImplementedError
-
-    async def _get_html(self, url: str, data: dict | None = None) -> str:
-        headers = self.headers
-        headers["Referer"] = url
-        headers["User-Agent"] = random.choice(USER_AGENTS)
-        if data:
-            async with (
-                ClientSession() as session,
-                session.post(
-                    url,
-                    headers=headers,
-                    data=data,
-                    timeout=self.TIMEOUT,
-                ) as resp,
-            ):
-                ret = await resp.text(encoding="utf-8")
-                return ret
-        else:
-            async with (
-                ClientSession() as session,
-                session.get(
-                    url,
-                    headers=headers,
-                    timeout=self.TIMEOUT,
-                ) as resp,
-            ):
-                ret = await resp.text(encoding="utf-8")
-                return ret
-
-    def tidy_text(self, text: str) -> str:
-        """清理文本，去除空格、换行符等"""
-        return text.strip().replace("\n", " ").replace("\r", " ").replace("  ", " ")
-
-    def _get_url(self, tag: Tag) -> str:
-        return self.tidy_text(tag.get_text())
-
-    async def search(self, query: str, num_results: int) -> list[SearchResult]:
-        query = urllib.parse.quote(query)
-
-        try:
-            resp = await self._get_next_page(query)
-            soup = BeautifulSoup(resp, "html.parser")
-            links = soup.select(self._set_selector("links"))
-            results = []
-            for link in links:
-                # Safely get the title text (select_one may return None)
-                title_elem = link.select_one(self._set_selector("title"))
-                title = ""
-                if title_elem is not None:
-                    title = self.tidy_text(title_elem.get_text())
-
-                url_tag = link.select_one(self._set_selector("url"))
-                snippet = ""
-                if title and url_tag:
-                    url = self._get_url(url_tag)
-                    results.append(SearchResult(title=title, url=url, snippet=snippet))
-            return results[:num_results] if len(results) > num_results else results
-        except Exception as e:
-            raise e

astrbot/builtin_stars/web_searcher/engines/bing.py

@@ -1,30 +0,0 @@
-from . import USER_AGENT_BING, SearchEngine
-
-
-class Bing(SearchEngine):
-    def __init__(self) -> None:
-        super().__init__()
-        self.base_urls = ["https://cn.bing.com", "https://www.bing.com"]
-        self.headers.update({"User-Agent": USER_AGENT_BING})
-
-    def _set_selector(self, selector: str):
-        selectors = {
-            "url": "div.b_attribution cite",
-            "title": "h2",
-            "text": "p",
-            "links": "ol#b_results > li.b_algo",
-            "next": 'div#b_content nav[role="navigation"] a.sb_pagN',
-        }
-        return selectors[selector]
-
-    async def _get_next_page(self, query) -> str:
-        # if self.page == 1:
-        #     await self._get_html(self.base_url)
-        for base_url in self.base_urls:
-            try:
-                url = f"{base_url}/search?q={query}"
-                return await self._get_html(url, None)
-            except Exception as _:
-                self.base_url = base_url
-                continue
-        raise Exception("Bing search failed")

astrbot/builtin_stars/web_searcher/engines/sogo.py

@@ -1,52 +0,0 @@
-import random
-import re
-from typing import cast
-
-from bs4 import BeautifulSoup, Tag
-
-from . import USER_AGENTS, SearchEngine, SearchResult
-
-
-class Sogo(SearchEngine):
-    def __init__(self) -> None:
-        super().__init__()
-        self.base_url = "https://www.sogou.com"
-        self.headers["User-Agent"] = random.choice(USER_AGENTS)
-
-    def _set_selector(self, selector: str):
-        selectors = {
-            "url": "h3 > a",
-            "title": "h3",
-            "text": "",
-            "links": "div.results > div.vrwrap:not(.middle-better-hintBox)",
-            "next": "",
-        }
-        return selectors[selector]
-
-    async def _get_next_page(self, query) -> str:
-        url = f"{self.base_url}/web?query={query}"
-        return await self._get_html(url, None)
-
-    def _get_url(self, tag: Tag) -> str:
-        return cast(str, tag.get("href"))
-
-    async def search(self, query: str, num_results: int) -> list[SearchResult]:
-        results = await super().search(query, num_results)
-        for result in results:
-            if result.url.startswith("/link?"):
-                result.url = self.base_url + result.url
-                result.url = await self._parse_url(result.url)
-        return results
-
-    async def _parse_url(self, url) -> str:
-        html = await self._get_html(url)
-        soup = BeautifulSoup(html, "html.parser")
-        script = soup.find("script")
-        if script:
-            script_text = (
-                script.string if script.string is not None else script.get_text()
-            )
-            match = re.search(r'window.location.replace\("(.+?)"\)', script_text)
-            if match:
-                url = match.group(1)
-        return url

astrbot/builtin_stars/web_searcher/main.py

@@ -1,611 +0,0 @@
-import asyncio
-import json
-import random
-import uuid
-
-import aiohttp
-from bs4 import BeautifulSoup
-from readability import Document
-
-from astrbot.api import AstrBotConfig, llm_tool, logger, sp, star
-from astrbot.api.event import AstrMessageEvent, filter
-from astrbot.api.provider import ProviderRequest
-from astrbot.core.provider.func_tool_manager import FunctionToolManager
-
-from .engines import HEADERS, USER_AGENTS, SearchResult
-from .engines.bing import Bing
-from .engines.sogo import Sogo
-
-
-class Main(star.Star):
-    TOOLS = [
-        "web_search",
-        "fetch_url",
-        "web_search_tavily",
-        "tavily_extract_web_page",
-        "web_search_bocha",
-    ]
-
-    def __init__(self, context: star.Context) -> None:
-        self.context = context
-        self.tavily_key_index = 0
-        self.tavily_key_lock = asyncio.Lock()
-
-        self.bocha_key_index = 0
-        self.bocha_key_lock = asyncio.Lock()
-
-        # 将 str 类型的 key 迁移至 list[str]，并保存
-        cfg = self.context.get_config()
-        provider_settings = cfg.get("provider_settings")
-        if provider_settings:
-            tavily_key = provider_settings.get("websearch_tavily_key")
-            if isinstance(tavily_key, str):
-                logger.info(
-                    "检测到旧版 websearch_tavily_key (字符串格式)，自动迁移为列表格式并保存。",
-                )
-                if tavily_key:
-                    provider_settings["websearch_tavily_key"] = [tavily_key]
-                else:
-                    provider_settings["websearch_tavily_key"] = []
-                cfg.save_config()
-
-            bocha_key = provider_settings.get("websearch_bocha_key")
-            if isinstance(bocha_key, str):
-                if bocha_key:
-                    provider_settings["websearch_bocha_key"] = [bocha_key]
-                else:
-                    provider_settings["websearch_bocha_key"] = []
-                cfg.save_config()
-
-        self.bing_search = Bing()
-        self.sogo_search = Sogo()
-        self.baidu_initialized = False
-
-    async def _tidy_text(self, text: str) -> str:
-        """清理文本，去除空格、换行符等"""
-        return text.strip().replace("\n", " ").replace("\r", " ").replace("  ", " ")
-
-    async def _get_from_url(self, url: str) -> str:
-        """获取网页内容"""
-        header = HEADERS
-        header.update({"User-Agent": random.choice(USER_AGENTS)})
-        async with aiohttp.ClientSession(trust_env=True) as session:
-            async with session.get(url, headers=header) as response:
-                html = await response.text(encoding="utf-8")
-                doc = Document(html)
-                ret = doc.summary(html_partial=True)
-                soup = BeautifulSoup(ret, "html.parser")
-                ret = await self._tidy_text(soup.get_text())
-                return ret
-
-    async def _process_search_result(
-        self,
-        result: SearchResult,
-        idx: int,
-        websearch_link: bool,
-    ) -> str:
-        """处理单个搜索结果"""
-        logger.info(f"web_searcher - scraping web: {result.title} - {result.url}")
-        try:
-            site_result = await self._get_from_url(result.url)
-        except BaseException:
-            site_result = ""
-        site_result = (
-            f"{site_result[:700]}..." if len(site_result) > 700 else site_result
-        )
-
-        header = f"{idx}. {result.title} "
-
-        if websearch_link and result.url:
-            header += result.url
-
-        return f"{header}\n{result.snippet}\n{site_result}\n\n"
-
-    async def _web_search_default(
-        self,
-        query,
-        num_results: int = 5,
-    ) -> list[SearchResult]:
-        results = []
-        try:
-            results = await self.bing_search.search(query, num_results)
-        except Exception as e:
-            logger.error(f"bing search error: {e}, try the next one...")
-        if len(results) == 0:
-            logger.debug("search bing failed")
-            try:
-                results = await self.sogo_search.search(query, num_results)
-            except Exception as e:
-                logger.error(f"sogo search error: {e}")
-        if len(results) == 0:
-            logger.debug("search sogo failed")
-            return []
-
-        return results
-
-    async def _get_tavily_key(self, cfg: AstrBotConfig) -> str:
-        """并发安全的从列表中获取并轮换Tavily API密钥。"""
-        tavily_keys = cfg.get("provider_settings", {}).get("websearch_tavily_key", [])
-        if not tavily_keys:
-            raise ValueError("错误：Tavily API密钥未在AstrBot中配置。")
-
-        async with self.tavily_key_lock:
-            key = tavily_keys[self.tavily_key_index]
-            self.tavily_key_index = (self.tavily_key_index + 1) % len(tavily_keys)
-            return key
-
-    async def _web_search_tavily(
-        self,
-        cfg: AstrBotConfig,
-        payload: dict,
-    ) -> list[SearchResult]:
-        """使用 Tavily 搜索引擎进行搜索"""
-        tavily_key = await self._get_tavily_key(cfg)
-        url = "https://api.tavily.com/search"
-        header = {
-            "Authorization": f"Bearer {tavily_key}",
-            "Content-Type": "application/json",
-        }
-        async with aiohttp.ClientSession(trust_env=True) as session:
-            async with session.post(
-                url,
-                json=payload,
-                headers=header,
-            ) as response:
-                if response.status != 200:
-                    reason = await response.text()
-                    raise Exception(
-                        f"Tavily web search failed: {reason}, status: {response.status}",
-                    )
-                data = await response.json()
-                results = []
-                for item in data.get("results", []):
-                    result = SearchResult(
-                        title=item.get("title"),
-                        url=item.get("url"),
-                        snippet=item.get("content"),
-                        favicon=item.get("favicon"),
-                    )
-                    results.append(result)
-                return results
-
-    async def _extract_tavily(self, cfg: AstrBotConfig, payload: dict) -> list[dict]:
-        """使用 Tavily 提取网页内容"""
-        tavily_key = await self._get_tavily_key(cfg)
-        url = "https://api.tavily.com/extract"
-        header = {
-            "Authorization": f"Bearer {tavily_key}",
-            "Content-Type": "application/json",
-        }
-        async with aiohttp.ClientSession(trust_env=True) as session:
-            async with session.post(
-                url,
-                json=payload,
-                headers=header,
-            ) as response:
-                if response.status != 200:
-                    reason = await response.text()
-                    raise Exception(
-                        f"Tavily web search failed: {reason}, status: {response.status}",
-                    )
-                data = await response.json()
-                results: list[dict] = data.get("results", [])
-                if not results:
-                    raise ValueError(
-                        "Error: Tavily web searcher does not return any results.",
-                    )
-                return results
-
-    @llm_tool(name="web_search")
-    async def search_from_search_engine(
-        self,
-        event: AstrMessageEvent,
-        query: str,
-        max_results: int = 5,
-    ) -> str:
-        """搜索网络以回答用户的问题。当用户需要搜索网络以获取即时性的信息时调用此工具。
-
-        Args:
-            query(string): 和用户的问题最相关的搜索关键词，用于在 Google 上搜索。
-            max_results(number): 返回的最大搜索结果数量，默认为 5。
-
-        """
-        logger.info(f"web_searcher - search_from_search_engine: {query}")
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        websearch_link = cfg["provider_settings"].get("web_search_link", False)
-
-        results = await self._web_search_default(query, max_results)
-        if not results:
-            return "Error: web searcher does not return any results."
-
-        tasks = []
-        for idx, result in enumerate(results, 1):
-            task = self._process_search_result(result, idx, websearch_link)
-            tasks.append(task)
-        processed_results = await asyncio.gather(*tasks, return_exceptions=True)
-        ret = ""
-        for processed_result in processed_results:
-            if isinstance(processed_result, BaseException):
-                logger.error(f"Error processing search result: {processed_result}")
-                continue
-            ret += processed_result
-
-        if websearch_link:
-            ret += "\n\n针对问题，请根据上面的结果分点总结，并且在结尾处附上对应内容的参考链接（如有）。"
-
-        return ret
-
-    async def ensure_baidu_ai_search_mcp(self, umo: str | None = None) -> None:
-        if self.baidu_initialized:
-            return
-        cfg = self.context.get_config(umo=umo)
-        key = cfg.get("provider_settings", {}).get(
-            "websearch_baidu_app_builder_key",
-            "",
-        )
-        if not key:
-            raise ValueError(
-                "Error: Baidu AI Search API key is not configured in AstrBot.",
-            )
-        func_tool_mgr = self.context.get_llm_tool_manager()
-        await func_tool_mgr.enable_mcp_server(
-            "baidu_ai_search",
-            config={
-                "transport": "sse",
-                "url": f"http://appbuilder.baidu.com/v2/ai_search/mcp/sse?api_key={key}",
-                "headers": {},
-                "timeout": 600,
-            },
-        )
-        self.baidu_initialized = True
-        logger.info("Successfully initialized Baidu AI Search MCP server.")
-
-    @llm_tool(name="fetch_url")
-    async def fetch_website_content(self, event: AstrMessageEvent, url: str) -> str:
-        """Fetch the content of a website with the given web url
-
-        Args:
-            url(string): The url of the website to fetch content from
-
-        """
-        resp = await self._get_from_url(url)
-        return resp
-
-    @llm_tool("web_search_tavily")
-    async def search_from_tavily(
-        self,
-        event: AstrMessageEvent,
-        query: str,
-        max_results: int = 7,
-        search_depth: str = "basic",
-        topic: str = "general",
-        days: int = 3,
-        time_range: str = "",
-        start_date: str = "",
-        end_date: str = "",
-    ) -> str:
-        """A web search tool that uses Tavily to search the web for relevant content.
-        Ideal for gathering current information, news, and detailed web content analysis.
-
-        Args:
-            query(string): Required. Search query.
-            max_results(number): Optional. The maximum number of results to return. Default is 7. Range is 5-20.
-            search_depth(string): Optional. The depth of the search, must be one of 'basic', 'advanced'. Default is "basic".
-            topic(string): Optional. The topic of the search, must be one of 'general', 'news'. Default is "general".
-            days(number): Optional. The number of days back from the current date to include in the search results. Please note that this feature is only available when using the 'news' search topic.
-            time_range(string): Optional. The time range back from the current date to include in the search results. This feature is available for both 'general' and 'news' search topics. Must be one of 'day', 'week', 'month', 'year'.
-            start_date(string): Optional. The start date for the search results in the format 'YYYY-MM-DD'.
-            end_date(string): Optional. The end date for the search results in the format 'YYYY-MM-DD'.
-
-        """
-        logger.info(f"web_searcher - search_from_tavily: {query}")
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        # websearch_link = cfg["provider_settings"].get("web_search_link", False)
-        if not cfg.get("provider_settings", {}).get("websearch_tavily_key", []):
-            raise ValueError("Error: Tavily API key is not configured in AstrBot.")
-
-        # build payload
-        payload = {"query": query, "max_results": max_results, "include_favicon": True}
-        if search_depth not in ["basic", "advanced"]:
-            search_depth = "basic"
-        payload["search_depth"] = search_depth
-
-        if topic not in ["general", "news"]:
-            topic = "general"
-        payload["topic"] = topic
-
-        if topic == "news":
-            payload["days"] = days
-
-        if time_range in ["day", "week", "month", "year"]:
-            payload["time_range"] = time_range
-        if start_date:
-            payload["start_date"] = start_date
-        if end_date:
-            payload["end_date"] = end_date
-
-        results = await self._web_search_tavily(cfg, payload)
-        if not results:
-            return "Error: Tavily web searcher does not return any results."
-
-        ret_ls = []
-        ref_uuid = str(uuid.uuid4())[:4]
-        for idx, result in enumerate(results, 1):
-            index = f"{ref_uuid}.{idx}"
-            ret_ls.append(
-                {
-                    "title": f"{result.title}",
-                    "url": f"{result.url}",
-                    "snippet": f"{result.snippet}",
-                    # TODO: do not need ref for non-webchat platform adapter
-                    "index": index,
-                }
-            )
-            if result.favicon:
-                sp.temporary_cache["_ws_favicon"][result.url] = result.favicon
-        # ret = "\n".join(ret_ls)
-        ret = json.dumps({"results": ret_ls}, ensure_ascii=False)
-        return ret
-
-    @llm_tool("tavily_extract_web_page")
-    async def tavily_extract_web_page(
-        self,
-        event: AstrMessageEvent,
-        url: str = "",
-        extract_depth: str = "basic",
-    ) -> str:
-        """Extract the content of a web page using Tavily.
-
-        Args:
-            url(string): Required. An URl to extract content from.
-            extract_depth(string): Optional. The depth of the extraction, must be one of 'basic', 'advanced'. Default is "basic".
-
-        """
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        if not cfg.get("provider_settings", {}).get("websearch_tavily_key", []):
-            raise ValueError("Error: Tavily API key is not configured in AstrBot.")
-
-        if not url:
-            raise ValueError("Error: url must be a non-empty string.")
-        if extract_depth not in ["basic", "advanced"]:
-            extract_depth = "basic"
-        payload = {
-            "urls": [url],
-            "extract_depth": extract_depth,
-        }
-        results = await self._extract_tavily(cfg, payload)
-        ret_ls = []
-        for result in results:
-            ret_ls.append(f"URL: {result.get('url', 'No URL')}")
-            ret_ls.append(f"Content: {result.get('raw_content', 'No content')}")
-        ret = "\n".join(ret_ls)
-        if not ret:
-            return "Error: Tavily web searcher does not return any results."
-        return ret
-
-    async def _get_bocha_key(self, cfg: AstrBotConfig) -> str:
-        """并发安全的从列表中获取并轮换BoCha API密钥。"""
-        bocha_keys = cfg.get("provider_settings", {}).get("websearch_bocha_key", [])
-        if not bocha_keys:
-            raise ValueError("错误：BoCha API密钥未在AstrBot中配置。")
-
-        async with self.bocha_key_lock:
-            key = bocha_keys[self.bocha_key_index]
-            self.bocha_key_index = (self.bocha_key_index + 1) % len(bocha_keys)
-            return key
-
-    async def _web_search_bocha(
-        self,
-        cfg: AstrBotConfig,
-        payload: dict,
-    ) -> list[SearchResult]:
-        """使用 BoCha 搜索引擎进行搜索"""
-        bocha_key = await self._get_bocha_key(cfg)
-        url = "https://api.bochaai.com/v1/web-search"
-        header = {
-            "Authorization": f"Bearer {bocha_key}",
-            "Content-Type": "application/json",
-        }
-        async with aiohttp.ClientSession(trust_env=True) as session:
-            async with session.post(
-                url,
-                json=payload,
-                headers=header,
-            ) as response:
-                if response.status != 200:
-                    reason = await response.text()
-                    raise Exception(
-                        f"BoCha web search failed: {reason}, status: {response.status}",
-                    )
-                data = await response.json()
-                data = data["data"]["webPages"]["value"]
-                results = []
-                for item in data:
-                    result = SearchResult(
-                        title=item.get("name"),
-                        url=item.get("url"),
-                        snippet=item.get("snippet"),
-                        favicon=item.get("siteIcon"),
-                    )
-                    results.append(result)
-                return results
-
-    @llm_tool("web_search_bocha")
-    async def search_from_bocha(
-        self,
-        event: AstrMessageEvent,
-        query: str,
-        freshness: str = "noLimit",
-        summary: bool = False,
-        include: str = "",
-        exclude: str = "",
-        count: int = 10,
-    ) -> str:
-        """
-        A web search tool based on Bocha Search API, used to retrieve web pages
-        related to the user's query.
-
-        Args:
-            query (string): Required. User's search query.
-
-            freshness (string): Optional. Specifies the time range of the search.
-                Supported values:
-                - "noLimit": No time limit (default, recommended).
-                - "oneDay": Within one day.
-                - "oneWeek": Within one week.
-                - "oneMonth": Within one month.
-                - "oneYear": Within one year.
-                - "YYYY-MM-DD..YYYY-MM-DD": Search within a specific date range.
-                  Example: "2025-01-01..2025-04-06".
-                - "YYYY-MM-DD": Search on a specific date.
-                  Example: "2025-04-06".
-                It is recommended to use "noLimit", as the search algorithm will
-                automatically optimize time relevance. Manually restricting the
-                time range may result in no search results.
-
-            summary (boolean): Optional. Whether to include a text summary
-                for each search result.
-                - True: Include summary.
-                - False: Do not include summary (default).
-
-            include (string): Optional. Specifies the domains to include in
-                the search. Multiple domains can be separated by "|" or ",".
-                A maximum of 100 domains is allowed.
-                Examples:
-                - "qq.com"
-                - "qq.com|m.163.com"
-
-            exclude (string): Optional. Specifies the domains to exclude from
-                the search. Multiple domains can be separated by "|" or ",".
-                A maximum of 100 domains is allowed.
-                Examples:
-                - "qq.com"
-                - "qq.com|m.163.com"
-
-            count (number): Optional. Number of search results to return.
-                - Range: 1–50
-                - Default: 10
-                The actual number of returned results may be less than the
-                specified count.
-        """
-        logger.info(f"web_searcher - search_from_bocha: {query}")
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        # websearch_link = cfg["provider_settings"].get("web_search_link", False)
-        if not cfg.get("provider_settings", {}).get("websearch_bocha_key", []):
-            raise ValueError("Error: BoCha API key is not configured in AstrBot.")
-
-        # build payload
-        payload = {
-            "query": query,
-            "count": count,
-        }
-
-        # freshness：时间范围
-        if freshness:
-            payload["freshness"] = freshness
-
-        # 是否返回摘要
-        payload["summary"] = summary
-
-        # include：限制搜索域
-        if include:
-            payload["include"] = include
-
-        # exclude：排除搜索域
-        if exclude:
-            payload["exclude"] = exclude
-
-        results = await self._web_search_bocha(cfg, payload)
-        if not results:
-            return "Error: BoCha web searcher does not return any results."
-
-        ret_ls = []
-        ref_uuid = str(uuid.uuid4())[:4]
-        for idx, result in enumerate(results, 1):
-            index = f"{ref_uuid}.{idx}"
-            ret_ls.append(
-                {
-                    "title": f"{result.title}",
-                    "url": f"{result.url}",
-                    "snippet": f"{result.snippet}",
-                    "index": index,
-                }
-            )
-            if result.favicon:
-                sp.temporary_cache["_ws_favicon"][result.url] = result.favicon
-        # ret = "\n".join(ret_ls)
-        ret = json.dumps({"results": ret_ls}, ensure_ascii=False)
-        return ret
-
-    @filter.on_llm_request(priority=-10000)
-    async def edit_web_search_tools(
-        self,
-        event: AstrMessageEvent,
-        req: ProviderRequest,
-    ) -> None:
-        """Get the session conversation for the given event."""
-        cfg = self.context.get_config(umo=event.unified_msg_origin)
-        prov_settings = cfg.get("provider_settings", {})
-        websearch_enable = prov_settings.get("web_search", False)
-        provider = prov_settings.get("websearch_provider", "default")
-
-        tool_set = req.func_tool
-        if isinstance(tool_set, FunctionToolManager):
-            req.func_tool = tool_set.get_full_tool_set()
-            tool_set = req.func_tool
-
-        if not tool_set:
-            return
-
-        if not websearch_enable:
-            # pop tools
-            for tool_name in self.TOOLS:
-                tool_set.remove_tool(tool_name)
-            return
-
-        func_tool_mgr = self.context.get_llm_tool_manager()
-        if provider == "default":
-            web_search_t = func_tool_mgr.get_func("web_search")
-            fetch_url_t = func_tool_mgr.get_func("fetch_url")
-            if web_search_t:
-                tool_set.add_tool(web_search_t)
-            if fetch_url_t:
-                tool_set.add_tool(fetch_url_t)
-            tool_set.remove_tool("web_search_tavily")
-            tool_set.remove_tool("tavily_extract_web_page")
-            tool_set.remove_tool("AIsearch")
-            tool_set.remove_tool("web_search_bocha")
-        elif provider == "tavily":
-            web_search_tavily = func_tool_mgr.get_func("web_search_tavily")
-            tavily_extract_web_page = func_tool_mgr.get_func("tavily_extract_web_page")
-            if web_search_tavily:
-                tool_set.add_tool(web_search_tavily)
-            if tavily_extract_web_page:
-                tool_set.add_tool(tavily_extract_web_page)
-            tool_set.remove_tool("web_search")
-            tool_set.remove_tool("fetch_url")
-            tool_set.remove_tool("AIsearch")
-            tool_set.remove_tool("web_search_bocha")
-        elif provider == "baidu_ai_search":
-            try:
-                await self.ensure_baidu_ai_search_mcp(event.unified_msg_origin)
-                aisearch_tool = func_tool_mgr.get_func("AIsearch")
-                if not aisearch_tool:
-                    raise ValueError("Cannot get Baidu AI Search MCP tool.")
-                tool_set.add_tool(aisearch_tool)
-                tool_set.remove_tool("web_search")
-                tool_set.remove_tool("fetch_url")
-                tool_set.remove_tool("web_search_tavily")
-                tool_set.remove_tool("tavily_extract_web_page")
-                tool_set.remove_tool("web_search_bocha")
-            except Exception as e:
-                logger.error(f"Cannot Initialize Baidu AI Search MCP Server: {e}")
-        elif provider == "bocha":
-            web_search_bocha = func_tool_mgr.get_func("web_search_bocha")
-            if web_search_bocha:
-                tool_set.add_tool(web_search_bocha)
-            tool_set.remove_tool("web_search")
-            tool_set.remove_tool("fetch_url")
-            tool_set.remove_tool("AIsearch")
-            tool_set.remove_tool("web_search_tavily")
-            tool_set.remove_tool("tavily_extract_web_page")

astrbot/builtin_stars/web_searcher/metadata.yaml

@@ -1,4 +0,0 @@
-name: astrbot-web-searcher
-desc: 让 LLM 具有网页检索能力
-author: Soulter
-version: 1.14.514

astrbot/cli/__init__.py

@@ -1 +1 @@
-__version__ = "4.20.0"
+__version__ = "4.23.0"

astrbot/core/agent/context/token_counter.py

@@ -1,7 +1,7 @@
 import json
 from typing import Protocol, runtime_checkable
 
-from ..message import Message, TextPart
+from ..message import AudioURLPart, ImageURLPart, Message, TextPart, ThinkPart
 
 
 @runtime_checkable
@@ -28,9 +28,19 @@ class TokenCounter(Protocol):
         ...
 
 
+# 图片/音频 token 开销估算值，参考 OpenAI vision pricing:
+# low-res ~85 tokens, high-res ~170 per 512px tile, 通常几百到上千。
+# 这里取一个保守中位数，宁可偏高触发压缩也不要偏低导致 API 报错。
+IMAGE_TOKEN_ESTIMATE = 765
+AUDIO_TOKEN_ESTIMATE = 500
+
+
 class EstimateTokenCounter:
     """Estimate token counter implementation.
     Provides a simple estimation of token count based on character types.
+
+    Supports multimodal content: images, audio, and thinking parts
+    are all counted so that the context compressor can trigger in time.
     """
 
     def count_tokens(
@@ -45,12 +55,16 @@ class EstimateTokenCounter:
             if isinstance(content, str):
                 total += self._estimate_tokens(content)
             elif isinstance(content, list):
-                # 处理多模态内容
                 for part in content:
                     if isinstance(part, TextPart):
                         total += self._estimate_tokens(part.text)
+                    elif isinstance(part, ThinkPart):
+                        total += self._estimate_tokens(part.think)
+                    elif isinstance(part, ImageURLPart):
+                        total += IMAGE_TOKEN_ESTIMATE
+                    elif isinstance(part, AudioURLPart):
+                        total += AUDIO_TOKEN_ESTIMATE
 
-            # 处理 Tool Calls
             if msg.tool_calls:
                 for tc in msg.tool_calls:
                     tc_str = json.dumps(tc if isinstance(tc, dict) else tc.model_dump())

astrbot/core/agent/context/truncator.py

@@ -12,14 +12,50 @@ class ContextTruncator:
             and len(message.tool_calls) > 0
         )
 
+    @staticmethod
+    def _split_system_rest(
+        messages: list[Message],
+    ) -> tuple[list[Message], list[Message]]:
+        """Split messages into system messages and the rest.
+
+        Returns:
+            tuple: (system_messages, non_system_messages)
+        """
+        first_non_system = 0
+        for i, msg in enumerate(messages):
+            if msg.role != "system":
+                first_non_system = i
+                break
+        return messages[:first_non_system], messages[first_non_system:]
+
+    @staticmethod
+    def _ensure_user_message(
+        system_messages: list[Message],
+        truncated: list[Message],
+        original_messages: list[Message],
+    ) -> list[Message]:
+        """Ensure the result always contains the first user message right after
+        system messages. This is required by many LLM APIs (e.g. Zhipu) that
+        mandate a ``user`` message immediately following the ``system`` message.
+        """
+        if truncated and truncated[0].role == "user":
+            return system_messages + truncated
+
+        # Locate the first user message from the *original* list.
+        first_user = next((m for m in original_messages if m.role == "user"), None)
+        if first_user is None:
+            return system_messages + truncated
+
+        return system_messages + [first_user] + truncated
+
     def fix_messages(self, messages: list[Message]) -> list[Message]:
-        """修复消息列表，确保 tool call 和 tool response 的配对关系有效。
+        """Fix the message list to ensure the validity of tool call and tool response pairing.
 
-        此方法确保：
-        1. 每个 `tool` 消息前面都有一个包含 tool_calls 的 `assistant` 消息
-        2. 每个包含 tool_calls 的 `assistant` 消息后面都有对应的 `tool` 响应
+        This method ensures that:
+        1. Each `tool` message is preceded by an `assistant` message containing `tool_calls`.
+        2. Each `assistant` message containing `tool_calls` is followed by corresponding `
 
-        这是 OpenAI Chat Completions API 规范的要求（Gemini 对此执行严格检查）。
+        This is a requirement of the OpenAI Chat Completions API specification (Gemini enforces this strictly).
         """
         if not messages:
             return messages
@@ -38,24 +74,25 @@ class ContextTruncator:
 
         for msg in messages:
             if msg.role == "tool":
-                # 只有在有挂起的 assistant(tool_calls) 时才记录 tool 响应
+                # Only record tool responses when there is a pending assistant(tool_calls)
                 if pending_assistant is not None:
                     pending_tools.append(msg)
-                # else: 孤立的 tool 消息，直接忽略
+                # Isolated tool messages without a preceding assistant(tool_calls) are ignored
                 continue
 
             if self._has_tool_calls(msg):
-                # 遇到新的 assistant(tool_calls) 前，先处理旧的 pending 链
+                # When encountering a new assistant(tool_calls), first process the old pending chain
                 flush_pending_if_valid()
                 pending_assistant = msg
                 continue
 
-            # 非 tool，且不含 tool_calls 的消息
-            # 先结束任何 pending 链，再正常追加
+            # Non-tool messages that do not contain tool_calls will break the pending chain.
+            # Flush any pending chain first, then append the current message normally.
             flush_pending_if_valid()
             fixed_messages.append(msg)
 
-        # 结束时处理最后一个 pending 链
+        # Flush the last pending chain at the end,
+        # ensuring that any remaining valid assistant(tool_calls) and its tools are included in the final list.
         flush_pending_if_valid()
 
         return fixed_messages
@@ -66,29 +103,23 @@ class ContextTruncator:
         keep_most_recent_turns: int,
         drop_turns: int = 1,
     ) -> list[Message]:
-        """截断上下文列表，确保不超过最大长度。
-        一个 turn 包含一个 user 消息和一个 assistant 消息。
-        这个方法会保证截断后的上下文列表符合 OpenAI 的上下文格式。
+        """
+        Turn-based truncation strategy, which drops the oldest turns while keeping the most recent N turns.
+        A turn consists of a user message and an assistant message.
+        This method ensures that the truncated context list conforms to OpenAI's context format.
 
         Args:
-            messages: 上下文列表
-            keep_most_recent_turns: 保留最近的对话轮数
-            drop_turns: 一次性丢弃的对话轮数
+            messages: The original list of messages in the context.
+            keep_most_recent_turns: The number of most recent turns to keep. If set to -1, it means keeping all turns (no truncation).
+            drop_turns: The number of turns to drop from the beginning.
 
         Returns:
-            截断后的上下文列表
+            The truncated list of messages.
         """
         if keep_most_recent_turns == -1:
             return messages
 
-        first_non_system = 0
-        for i, msg in enumerate(messages):
-            if msg.role != "system":
-                first_non_system = i
-                break
-
-        system_messages = messages[:first_non_system]
-        non_system_messages = messages[first_non_system:]
+        system_messages, non_system_messages = self._split_system_rest(messages)
 
         if len(non_system_messages) // 2 <= keep_most_recent_turns:
             return messages
@@ -99,7 +130,7 @@ class ContextTruncator:
         else:
             truncated_contexts = non_system_messages[-num_to_keep * 2 :]
 
-        # 找到第一个 role 为 user 的索引，确保上下文格式正确
+        # Find the first user message
         index = next(
             (i for i, item in enumerate(truncated_contexts) if item.role == "user"),
             None,
@@ -107,8 +138,9 @@ class ContextTruncator:
         if index is not None and index > 0:
             truncated_contexts = truncated_contexts[index:]
 
-        result = system_messages + truncated_contexts
-
+        result = self._ensure_user_message(
+            system_messages, truncated_contexts, messages
+        )
         return self.fix_messages(result)
 
     def truncate_by_dropping_oldest_turns(
@@ -116,53 +148,39 @@ class ContextTruncator:
         messages: list[Message],
         drop_turns: int = 1,
     ) -> list[Message]:
-        """丢弃最旧的 N 个对话轮次。"""
+        """Drop the oldest N turns, regardless of the number of turns to keep."""
         if drop_turns <= 0:
             return messages
 
-        first_non_system = 0
-        for i, msg in enumerate(messages):
-            if msg.role != "system":
-                first_non_system = i
-                break
-
-        system_messages = messages[:first_non_system]
-        non_system_messages = messages[first_non_system:]
+        system_messages, non_system_messages = self._split_system_rest(messages)
 
         if len(non_system_messages) // 2 <= drop_turns:
             truncated_non_system = []
         else:
             truncated_non_system = non_system_messages[drop_turns * 2 :]
 
+        # Find the first user message
         index = next(
             (i for i, item in enumerate(truncated_non_system) if item.role == "user"),
             None,
         )
         if index is not None:
             truncated_non_system = truncated_non_system[index:]
-        elif truncated_non_system:
-            truncated_non_system = []
-
-        result = system_messages + truncated_non_system
 
+        result = self._ensure_user_message(
+            system_messages, truncated_non_system, messages
+        )
         return self.fix_messages(result)
 
     def truncate_by_halving(
         self,
         messages: list[Message],
     ) -> list[Message]:
-        """对半砍策略，删除 50% 的消息"""
+        """Halve the number of messages, keeping the most recent ones."""
         if len(messages) <= 2:
             return messages
 
-        first_non_system = 0
-        for i, msg in enumerate(messages):
-            if msg.role != "system":
-                first_non_system = i
-                break
-
-        system_messages = messages[:first_non_system]
-        non_system_messages = messages[first_non_system:]
+        system_messages, non_system_messages = self._split_system_rest(messages)
 
         messages_to_delete = len(non_system_messages) // 2
         if messages_to_delete == 0:
@@ -170,6 +188,7 @@ class ContextTruncator:
 
         truncated_non_system = non_system_messages[messages_to_delete:]
 
+        # Find the first user message
         index = next(
             (i for i, item in enumerate(truncated_non_system) if item.role == "user"),
             None,
@@ -177,6 +196,7 @@ class ContextTruncator:
         if index is not None:
             truncated_non_system = truncated_non_system[index:]
 
-        result = system_messages + truncated_non_system
-
+        result = self._ensure_user_message(
+            system_messages, truncated_non_system, messages
+        )
         return self.fix_messages(result)

astrbot/core/agent/handoff.py

@@ -15,7 +15,6 @@ class HandoffTool(FunctionTool, Generic[TContext]):
         tool_description: str | None = None,
         **kwargs,
     ) -> None:
-
         # Avoid passing duplicate `description` to the FunctionTool dataclass.
         # Some call sites (e.g. SubAgentOrchestrator) pass `description` via kwargs
         # to override what the main agent sees, while we also compute a default
@@ -62,4 +61,4 @@ class HandoffTool(FunctionTool, Generic[TContext]):
 
     def default_description(self, agent_name: str | None) -> str:
         agent_name = agent_name or "another"
-        return f"Delegate tasks to {self.name} agent to handle the request."
+        return f"Delegate tasks to {agent_name} agent to handle the request."

astrbot/core/agent/mcp_client.py

@@ -1,7 +1,11 @@
 import asyncio
 import logging
+import os
+import re
+import sys
 from contextlib import AsyncExitStack
 from datetime import timedelta
+from pathlib import Path, PureWindowsPath
 from typing import Generic
 
 from tenacity import (
@@ -19,6 +23,75 @@ from astrbot.core.utils.log_pipe import LogPipe
 from .run_context import TContext
 from .tool import FunctionTool
 
+_DEFAULT_STDIO_COMMAND_ALLOWLIST = frozenset(
+    {
+        "python",
+        "python3",
+        "py",
+        "node",
+        "npx",
+        "npm",
+        "pnpm",
+        "yarn",
+        "bun",
+        "bunx",
+        "deno",
+        "uv",
+        "uvx",
+    }
+)
+_DENIED_STDIO_COMMANDS = frozenset(
+    {
+        "bash",
+        "sh",
+        "zsh",
+        "fish",
+        "cmd",
+        "cmd.exe",
+        "powershell",
+        "powershell.exe",
+        "pwsh",
+        "pwsh.exe",
+        "osascript",
+        "open",
+        "curl",
+        "wget",
+        "nc",
+        "netcat",
+        "telnet",
+        "ssh",
+        "scp",
+        "rm",
+        "mv",
+        "cp",
+        "dd",
+        "mkfs",
+        "sudo",
+        "su",
+        "chmod",
+        "chown",
+        "kill",
+        "killall",
+        "shutdown",
+        "reboot",
+        "poweroff",
+        "halt",
+    }
+)
+_SHELL_META_RE = re.compile(r"[\r\n\x00;&|<>`$]")
+_PYTHON_INLINE_CODE_FLAGS = frozenset({"-c"})
+_JS_INLINE_CODE_FLAGS = frozenset({"-e", "--eval", "-p", "--print"})
+_DENIED_DOCKER_ARGS = frozenset(
+    {
+        "--privileged",
+        "--pid=host",
+        "--network=host",
+        "--net=host",
+        "--ipc=host",
+    }
+)
+_STDIO_ALLOWLIST_ENV = "ASTRBOT_MCP_STDIO_ALLOWED_COMMANDS"
+
 try:
     import anyio
     import mcp
@@ -40,11 +113,156 @@ def _prepare_config(config: dict) -> dict:
     """Prepare configuration, handle nested format"""
     if config.get("mcpServers"):
         first_key = next(iter(config["mcpServers"]))
-        config = config["mcpServers"][first_key]
+        config = dict(config["mcpServers"][first_key])
+    else:
+        config = dict(config)
     config.pop("active", None)
     return config
 
 
+def _normalize_stdio_command_name(command: str) -> str:
+    command = command.strip()
+    if "\\" in command:
+        command_name = PureWindowsPath(command).name
+    else:
+        command_name = Path(command).name
+    command_name = command_name.lower()
+    for suffix in (".exe", ".cmd", ".bat"):
+        if command_name.endswith(suffix):
+            return command_name[: -len(suffix)]
+    return command_name
+
+
+def _get_stdio_command_allowlist() -> set[str]:
+    allowed = set(_DEFAULT_STDIO_COMMAND_ALLOWLIST)
+    configured = os.environ.get(_STDIO_ALLOWLIST_ENV, "")
+    if configured.strip():
+        allowed = {
+            _normalize_stdio_command_name(item)
+            for item in configured.split(",")
+            if item.strip()
+        }
+    return allowed
+
+
+def _is_stdio_config(config: dict) -> bool:
+    cfg = _prepare_config(config.copy())
+    return "url" not in cfg
+
+
+def _validate_stdio_args(command_name: str, args: object) -> None:
+    if args is None:
+        return
+    if not isinstance(args, list) or not all(isinstance(arg, str) for arg in args):
+        raise ValueError("MCP stdio args must be a list of strings.")
+
+    for arg in args:
+        if "\x00" in arg or "\r" in arg or "\n" in arg:
+            raise ValueError("MCP stdio args cannot contain control characters.")
+
+    if command_name.startswith("python") or command_name == "py":
+        if any(
+            arg == "-c"
+            or (arg.startswith("-") and not arg.startswith("--") and "c" in arg)
+            for arg in args
+        ):
+            raise ValueError(
+                "MCP stdio Python servers must be launched from a module or file; inline code flags such as -c are not allowed."
+            )
+    elif command_name in {"node", "deno", "bun"} or command_name.startswith("node"):
+        if any(
+            arg in _JS_INLINE_CODE_FLAGS
+            or arg == "eval"
+            or (
+                arg.startswith("-")
+                and not arg.startswith("--")
+                and any(c in arg for c in "ep")
+            )
+            for arg in args
+        ):
+            raise ValueError(
+                "MCP stdio JavaScript servers must be launched from a package or file; inline eval flags are not allowed."
+            )
+    elif command_name == "docker":
+        denied = []
+        for i, arg in enumerate(args):
+            if arg in _DENIED_DOCKER_ARGS:
+                denied.append(arg)
+            elif (
+                arg in {"--network", "--net", "--pid", "--ipc"}
+                and i + 1 < len(args)
+                and args[i + 1] == "host"
+            ):
+                denied.append(f"{arg} {args[i + 1]}")
+        if denied:
+            raise ValueError(
+                f"MCP stdio Docker args are unsafe and not allowed: {', '.join(denied)}."
+            )
+
+
+def validate_mcp_stdio_config(config: dict) -> None:
+    """Validate stdio MCP config before any subprocess can be spawned."""
+    cfg = _prepare_config(config.copy())
+    if "url" in cfg:
+        return
+
+    command = cfg.get("command")
+    if not isinstance(command, str) or not command.strip():
+        raise ValueError("MCP stdio server requires a non-empty command.")
+    if _SHELL_META_RE.search(command):
+        raise ValueError("MCP stdio command contains unsafe shell metacharacters.")
+
+    command_name = _normalize_stdio_command_name(command)
+    if command_name in _DENIED_STDIO_COMMANDS:
+        raise ValueError(f"MCP stdio command `{command_name}` is not allowed.")
+
+    allowed = _get_stdio_command_allowlist()
+    if command_name not in allowed:
+        allowed_display = ", ".join(sorted(allowed))
+        raise ValueError(
+            f"MCP stdio command `{command_name}` is not allowed. "
+            f"Allowed commands: {allowed_display}. "
+            f"Set {_STDIO_ALLOWLIST_ENV} to override this list if you trust another launcher."
+        )
+
+    _validate_stdio_args(command_name, cfg.get("args"))
+
+    env = cfg.get("env")
+    if env is not None and not isinstance(env, dict):
+        raise ValueError("MCP stdio env must be an object.")
+    if isinstance(env, dict) and not all(
+        isinstance(key, str) and isinstance(value, str) for key, value in env.items()
+    ):
+        raise ValueError("MCP stdio env keys and values must be strings.")
+
+
+def _prepare_stdio_env(config: dict) -> dict:
+    """Preserve Windows executable resolution for stdio subprocesses."""
+    if sys.platform != "win32":
+        return config
+    prepared = config.copy()
+    env = dict(prepared.get("env") or {})
+    env = _merge_environment_variables(env)
+    prepared["env"] = env
+    return prepared
+
+
+def _merge_environment_variables(env: dict) -> dict:
+    """合并环境变量，处理Windows不区分大小写的情况"""
+    merged = env.copy()
+
+    # 将用户环境变量转换为统一的大小写形式便于比较
+    user_keys_lower = {k.lower(): k for k in merged.keys()}
+
+    for sys_key, sys_value in os.environ.items():
+        sys_key_lower = sys_key.lower()
+        if sys_key_lower not in user_keys_lower:
+            # 使用系统环境变量中的原始大小写
+            merged[sys_key] = sys_value
+
+    return merged
+
+
 async def _quick_test_mcp_connection(config: dict) -> tuple[bool, str]:
     """Quick test MCP server connectivity"""
     import aiohttp
@@ -214,6 +432,8 @@ class MCPClient:
                 )
 
         else:
+            validate_mcp_stdio_config(cfg)
+            cfg = _prepare_stdio_env(cfg)
             server_params = mcp.StdioServerParameters(
                 **cfg,
             )

astrbot/core/agent/run_context.py

@@ -16,7 +16,7 @@ class ContextWrapper(Generic[TContext]):
     context: TContext
     messages: list[Message] = Field(default_factory=list)
     """This field stores the llm message context for the agent run, agent runners will maintain this field automatically."""
-    tool_call_timeout: int = 60  # Default tool call timeout in seconds
+    tool_call_timeout: int = 120  # Default tool call timeout in seconds
 
 
 NoContext = ContextWrapper[None]

astrbot/core/agent/runners/deerflow/deerflow_agent_runner.py

@@ -410,18 +410,20 @@ class DeerFlowAgentRunner(BaseAgentRunner[TContext]):
         )
         return messages
 
-    def _build_runtime_context(self, thread_id: str) -> dict[str, T.Any]:
-        runtime_context: dict[str, T.Any] = {
+    def _build_runtime_configurable(self, thread_id: str) -> dict[str, T.Any]:
+        runtime_configurable: dict[str, T.Any] = {
             "thread_id": thread_id,
             "thinking_enabled": self.thinking_enabled,
             "is_plan_mode": self.plan_mode,
             "subagent_enabled": self.subagent_enabled,
         }
         if self.subagent_enabled:
-            runtime_context["max_concurrent_subagents"] = self.max_concurrent_subagents
+            runtime_configurable["max_concurrent_subagents"] = (
+                self.max_concurrent_subagents
+            )
         if self.model_name:
-            runtime_context["model_name"] = self.model_name
-        return runtime_context
+            runtime_configurable["model_name"] = self.model_name
+        return runtime_configurable
 
     def _build_payload(
         self,
@@ -430,16 +432,19 @@ class DeerFlowAgentRunner(BaseAgentRunner[TContext]):
         image_urls: list[str],
         system_prompt: str | None,
     ) -> dict[str, T.Any]:
+        runtime_configurable = self._build_runtime_configurable(thread_id)
         return {
             "assistant_id": self.assistant_id,
             "input": {
                 "messages": self._build_messages(prompt, image_urls, system_prompt),
             },
             "stream_mode": ["values", "messages-tuple", "custom"],
-            # LangGraph 0.6+ prefers context instead of configurable.
-            "context": self._build_runtime_context(thread_id),
+            # DeerFlow 2.0 consumes runtime overrides from config.configurable.
+            # Keep the legacy context mirror for older compat paths.
+            "context": dict(runtime_configurable),
             "config": {
                 "recursion_limit": self.recursion_limit,
+                "configurable": runtime_configurable,
             },
         }
 

astrbot/core/agent/runners/deerflow/deerflow_api_client.py

@@ -10,6 +10,33 @@ from astrbot.core import logger
 SSE_MAX_BUFFER_CHARS = 1_048_576
 
 
+class DeerFlowAPIError(Exception):
+    def __init__(
+        self,
+        *,
+        operation: str,
+        status: int,
+        body: str,
+        url: str,
+        thread_id: str | None = None,
+    ) -> None:
+        self.operation = operation
+        self.status = status
+        self.body = body
+        self.url = url
+        self.thread_id = thread_id
+
+        message = (
+            f"DeerFlow {operation} failed: status={status}, url={url}, body={body}"
+        )
+        if thread_id is not None:
+            message = (
+                f"DeerFlow {operation} failed: thread_id={thread_id}, "
+                f"status={status}, url={url}, body={body}"
+            )
+        super().__init__(message)
+
+
 def _normalize_sse_newlines(text: str) -> str:
     """Normalize CRLF/CR to LF so SSE block splitting works reliably."""
     return text.replace("\r\n", "\n").replace("\r", "\n")
@@ -152,11 +179,33 @@ class DeerFlowAPIClient:
         ) as resp:
             if resp.status not in (200, 201):
                 text = await resp.text()
-                raise Exception(
-                    f"DeerFlow create thread failed: {resp.status}. {text}",
+                raise DeerFlowAPIError(
+                    operation="create thread",
+                    status=resp.status,
+                    body=text,
+                    url=url,
                 )
             return await resp.json()
 
+    async def delete_thread(self, thread_id: str, timeout: float = 20) -> None:
+        session = self._get_session()
+        url = f"{self.api_base}/api/threads/{thread_id}"
+        async with session.delete(
+            url,
+            headers=self.headers,
+            timeout=timeout,
+            proxy=self.proxy,
+        ) as resp:
+            if resp.status not in (200, 202, 204, 404):
+                text = await resp.text()
+                raise DeerFlowAPIError(
+                    operation="delete thread",
+                    status=resp.status,
+                    body=text,
+                    url=url,
+                    thread_id=thread_id,
+                )
+
     async def stream_run(
         self,
         thread_id: str,
@@ -200,8 +249,12 @@ class DeerFlowAPIClient:
         ) as resp:
             if resp.status != 200:
                 text = await resp.text()
-                raise Exception(
-                    f"DeerFlow runs/stream request failed: {resp.status}. {text}",
+                raise DeerFlowAPIError(
+                    operation="runs/stream request",
+                    status=resp.status,
+                    body=text,
+                    url=url,
+                    thread_id=thread_id,
                 )
             async for event in _stream_sse(resp):
                 yield event

astrbot/core/agent/runners/tool_loop_agent_runner.py

@@ -4,7 +4,11 @@ import sys
 import time
 import traceback
 import typing as T
+import uuid
+from collections.abc import AsyncIterator
+from contextlib import suppress
 from dataclasses import dataclass, field
+from pathlib import Path
 
 from mcp.types import (
     BlobResourceContents,
@@ -14,11 +18,18 @@ from mcp.types import (
     TextContent,
     TextResourceContents,
 )
+from tenacity import (
+    AsyncRetrying,
+    retry_if_exception_type,
+    stop_after_attempt,
+    wait_exponential,
+)
 
 from astrbot import logger
 from astrbot.core.agent.message import ImageURLPart, TextPart, ThinkPart
-from astrbot.core.agent.tool import ToolSet
+from astrbot.core.agent.tool import FunctionTool, ToolSet
 from astrbot.core.agent.tool_image_cache import tool_image_cache
+from astrbot.core.exceptions import EmptyModelOutputError
 from astrbot.core.message.components import Json
 from astrbot.core.message.message_event_result import (
     MessageChain,
@@ -36,7 +47,7 @@ from astrbot.core.provider.provider import Provider
 from ..context.compressor import ContextCompressor
 from ..context.config import ContextConfig
 from ..context.manager import ContextManager
-from ..context.token_counter import TokenCounter
+from ..context.token_counter import EstimateTokenCounter, TokenCounter
 from ..hooks import BaseAgentRunHooks
 from ..message import AssistantMessageSegment, Message, ToolCallMessageSegment
 from ..response import AgentResponseData, AgentStats
@@ -80,12 +91,108 @@ class FollowUpTicket:
     resolved: asyncio.Event = field(default_factory=asyncio.Event)
 
 
+class _ToolExecutionInterrupted(Exception):
+    """Raised when a running tool call is interrupted by a stop request."""
+
+
+ToolExecutorResultT = T.TypeVar("ToolExecutorResultT")
+
+
 class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
+    TOOL_RESULT_MAX_ESTIMATED_TOKENS = 27_500
+    TOOL_RESULT_PREVIEW_MAX_ESTIMATED_TOKENS = 7000
+    EMPTY_OUTPUT_RETRY_ATTEMPTS = 3
+    EMPTY_OUTPUT_RETRY_WAIT_MIN_S = 1
+    EMPTY_OUTPUT_RETRY_WAIT_MAX_S = 4
+    USER_INTERRUPTION_MESSAGE = (
+        "[SYSTEM: User actively interrupted the response generation. "
+        "Partial output before interruption is preserved.]"
+    )
+    FOLLOW_UP_NOTICE_TEMPLATE = (
+        "\n\n[SYSTEM NOTICE] User sent follow-up messages while tool execution "
+        "was in progress. Prioritize these follow-up instructions in your next "
+        "actions. In your very next action, briefly acknowledge to the user "
+        "that their follow-up message(s) were received before continuing.\n"
+        "{follow_up_lines}"
+    )
+    MAX_STEPS_REACHED_PROMPT = (
+        "Maximum tool call limit reached. "
+        "Stop calling tools, and based on the information you have gathered, "
+        "summarize your task and findings, and reply to the user directly."
+    )
+    SKILLS_LIKE_REQUERY_INSTRUCTION_TEMPLATE = (
+        "You have decided to call tool(s): {tool_names}. Now call the tool(s) "
+        "with required arguments using the tool schema, and follow the existing "
+        "tool-use rules."
+    )
+    SKILLS_LIKE_REQUERY_REPAIR_INSTRUCTION = (
+        "This is the second-stage tool execution step. "
+        "You must do exactly one of the following: "
+        "1. Call one of the selected tools using the provided tool schema. "
+        "2. If calling a tool is no longer possible or appropriate, reply to the user "
+        "with a brief explanation of why. "
+        "Do not return an empty response. "
+        "Do not ignore the selected tools without explanation."
+    )
+    REPEATED_TOOL_NOTICE_L1_THRESHOLD = 3
+    REPEATED_TOOL_NOTICE_L2_THRESHOLD = 4
+    REPEATED_TOOL_NOTICE_L3_THRESHOLD = 5
+    REPEATED_TOOL_NOTICE_L1_TEMPLATE = (
+        "\n\n[SYSTEM NOTICE] By the way, you have executed the same tool "
+        "`{tool_name}` {streak} times consecutively. Double-check whether another "
+        "tool, different arguments, or a summary would move the task forward better."
+    )
+    REPEATED_TOOL_NOTICE_L2_TEMPLATE = (
+        "\n\n[SYSTEM NOTICE] Important: you have executed the same tool "
+        "`{tool_name}` {streak} times consecutively. Unless this repetition is "
+        "clearly necessary, stop repeating the same action and either switch "
+        "tools, refine parameters, or summarize what is still missing."
+    )
+    REPEATED_TOOL_NOTICE_L3_TEMPLATE = (
+        "\n\n[SYSTEM NOTICE] Important: you have executed the same tool "
+        "`{tool_name}` {streak} times consecutively. Repetition is now very "
+        "high. Continue only if each call is clearly producing new information. "
+        "Otherwise, change strategy, adjust arguments, or explain the limitation "
+        "to the user."
+    )
+    TOOL_RESULT_OVERFLOW_NOTICE_TEMPLATE = (
+        "Truncated tool output preview shown above. "
+        "The tool output was too large to include directly and was written to "
+        "`{overflow_path}`. Use {read_tool_hint} to inspect it. "
+        "Use a narrower window when reading large files."
+    )
+
     def _get_persona_custom_error_message(self) -> str | None:
         """Read persona-level custom error message from event extras when available."""
         event = getattr(self.run_context.context, "event", None)
         return extract_persona_custom_error_message_from_event(event)
 
+    async def _complete_with_assistant_response(self, llm_resp: LLMResponse) -> None:
+        """Finalize the current step as a plain assistant response with no tool calls."""
+        self.final_llm_resp = llm_resp
+        self._transition_state(AgentState.DONE)
+        self.stats.end_time = time.time()
+
+        parts = []
+        if llm_resp.reasoning_content or llm_resp.reasoning_signature:
+            parts.append(
+                ThinkPart(
+                    think=llm_resp.reasoning_content,
+                    encrypted=llm_resp.reasoning_signature,
+                )
+            )
+        if llm_resp.completion_text:
+            parts.append(TextPart(text=llm_resp.completion_text))
+        if len(parts) == 0:
+            logger.warning("LLM returned empty assistant message with no tool calls.")
+        self.run_context.messages.append(Message(role="assistant", content=parts))
+
+        try:
+            await self.agent_hooks.on_agent_done(self.run_context, llm_resp)
+        except Exception as e:
+            logger.error(f"Error in on_agent_done hook: {e}", exc_info=True)
+        self._resolve_unconsumed_follow_ups()
+
     @override
     async def reset(
         self,
@@ -109,6 +216,8 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         custom_compressor: ContextCompressor | None = None,
         tool_schema_mode: str | None = "full",
         fallback_providers: list[Provider] | None = None,
+        tool_result_overflow_dir: str | None = None,
+        read_tool: FunctionTool | None = None,
         **kwargs: T.Any,
     ) -> None:
         self.req = request
@@ -120,6 +229,9 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         self.truncate_turns = truncate_turns
         self.custom_token_counter = custom_token_counter
         self.custom_compressor = custom_compressor
+        self.tool_result_overflow_dir = tool_result_overflow_dir
+        self.read_tool = read_tool
+        self._tool_result_token_counter = EstimateTokenCounter()
         # we will do compress when:
         # 1. before requesting LLM
         # TODO: 2. after LLM output a tool call
@@ -154,10 +266,12 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         self.tool_executor = tool_executor
         self.agent_hooks = agent_hooks
         self.run_context = run_context
-        self._stop_requested = False
         self._aborted = False
+        self._abort_signal = asyncio.Event()
         self._pending_follow_ups: list[FollowUpTicket] = []
         self._follow_up_seq = 0
+        self._last_tool_name: str | None = None
+        self._same_tool_streak = 0
 
         # These two are used for tool schema mode handling
         # We now have two modes:
@@ -199,6 +313,103 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         self.stats = AgentStats()
         self.stats.start_time = time.time()
 
+    def _read_tool_hint(self) -> str:
+        if self.read_tool is not None:
+            return f"`{self.read_tool.name}`"
+        return "the available file-read tool"
+
+    async def _write_tool_result_overflow_file(
+        self,
+        *,
+        tool_call_id: str,
+        content: str,
+    ) -> str:
+        if self.tool_result_overflow_dir is None:
+            raise ValueError("tool_result_overflow_dir is not configured")
+
+        overflow_dir = Path(self.tool_result_overflow_dir).resolve(strict=False)
+        safe_tool_call_id = (
+            "".join(
+                ch if ch.isalnum() or ch in {"-", "_", "."} else "_"
+                for ch in tool_call_id
+            ).strip("._")
+            or "tool_call"
+        )
+        file_name = f"{safe_tool_call_id}_{uuid.uuid4().hex[:8]}.txt"
+        overflow_path = overflow_dir / file_name
+
+        def _run() -> str:
+            overflow_dir.mkdir(parents=True, exist_ok=True)
+            overflow_path.write_text(content, encoding="utf-8")
+            return str(overflow_path)
+
+        return await asyncio.to_thread(_run)
+
+    async def _materialize_large_tool_result(
+        self,
+        *,
+        tool_call_id: str,
+        content: str,
+    ) -> str:
+        if self.tool_result_overflow_dir is None or self.read_tool is None:
+            return content
+
+        estimated_tokens = self._tool_result_token_counter.count_tokens(
+            [Message(role="tool", content=content, tool_call_id=tool_call_id)]
+        )
+        if estimated_tokens <= self.TOOL_RESULT_MAX_ESTIMATED_TOKENS:
+            return content
+
+        preview = self._truncate_tool_result_preview(content, tool_call_id=tool_call_id)
+        try:
+            overflow_path = await self._write_tool_result_overflow_file(
+                tool_call_id=tool_call_id,
+                content=content,
+            )
+        except Exception as exc:
+            logger.warning(
+                "Failed to spill oversized tool result for %s: %s",
+                tool_call_id,
+                exc,
+                exc_info=True,
+            )
+            error_notice = (
+                "Tool output exceeded the inline result limit "
+                f"({estimated_tokens} estimated tokens > "
+                f"{self.TOOL_RESULT_MAX_ESTIMATED_TOKENS}) and could not be written "
+                f"to `{self.tool_result_overflow_dir}`: {exc}"
+            )
+            if not preview:
+                return error_notice
+            return f"{preview}\n\n{error_notice}"
+
+        notice = self.TOOL_RESULT_OVERFLOW_NOTICE_TEMPLATE.format(
+            overflow_path=overflow_path,
+            read_tool_hint=self._read_tool_hint(),
+        )
+        if not preview:
+            return notice
+        return f"{preview}\n\n{notice}"
+
+    def _truncate_tool_result_preview(
+        self,
+        content: str,
+        *,
+        tool_call_id: str,
+    ) -> str:
+        preview = content
+        while preview:
+            estimated_tokens = self._tool_result_token_counter.count_tokens(
+                [Message(role="tool", content=preview, tool_call_id=tool_call_id)]
+            )
+            if estimated_tokens <= self.TOOL_RESULT_PREVIEW_MAX_ESTIMATED_TOKENS:
+                return preview
+            next_len = len(preview) // 2
+            if next_len <= 0:
+                break
+            preview = preview[:next_len]
+        return preview
+
     async def _iter_llm_responses(
         self, *, include_model: bool = True
     ) -> T.AsyncGenerator[LLMResponse, None]:
@@ -208,6 +419,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             "func_tool": self.req.func_tool,
             "session_id": self.req.session_id,
             "extra_user_content_parts": self.req.extra_user_content_parts,  # list[ContentPart]
+            "abort_signal": self._abort_signal,
         }
         if include_model:
             # For primary provider we keep explicit model selection if provided.
@@ -238,31 +450,61 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                     candidate_id,
                 )
             self.provider = candidate
-            has_stream_output = False
             try:
-                async for resp in self._iter_llm_responses(include_model=idx == 0):
-                    if resp.is_chunk:
-                        has_stream_output = True
-                        yield resp
-                        continue
+                retrying = AsyncRetrying(
+                    retry=retry_if_exception_type(EmptyModelOutputError),
+                    stop=stop_after_attempt(self.EMPTY_OUTPUT_RETRY_ATTEMPTS),
+                    wait=wait_exponential(
+                        multiplier=1,
+                        min=self.EMPTY_OUTPUT_RETRY_WAIT_MIN_S,
+                        max=self.EMPTY_OUTPUT_RETRY_WAIT_MAX_S,
+                    ),
+                    reraise=True,
+                )
 
-                    if (
-                        resp.role == "err"
-                        and not has_stream_output
-                        and (not is_last_candidate)
-                    ):
-                        last_err_response = resp
-                        logger.warning(
-                            "Chat Model %s returns error response, trying fallback to next provider.",
-                            candidate_id,
-                        )
-                        break
+                async for attempt in retrying:
+                    has_stream_output = False
+                    with attempt:
+                        try:
+                            async for resp in self._iter_llm_responses(
+                                include_model=idx == 0
+                            ):
+                                if resp.is_chunk:
+                                    has_stream_output = True
+                                    yield resp
+                                    continue
 
-                    yield resp
-                    return
+                                if (
+                                    resp.role == "err"
+                                    and not has_stream_output
+                                    and (not is_last_candidate)
+                                ):
+                                    last_err_response = resp
+                                    logger.warning(
+                                        "Chat Model %s returns error response, trying fallback to next provider.",
+                                        candidate_id,
+                                    )
+                                    break
 
-                if has_stream_output:
-                    return
+                                yield resp
+                                return
+
+                            if has_stream_output:
+                                return
+                        except EmptyModelOutputError:
+                            if has_stream_output:
+                                logger.warning(
+                                    "Chat Model %s returned empty output after streaming started; skipping empty-output retry.",
+                                    candidate_id,
+                                )
+                            else:
+                                logger.warning(
+                                    "Chat Model %s returned empty output on attempt %s/%s.",
+                                    candidate_id,
+                                    attempt.retry_state.attempt_number,
+                                    self.EMPTY_OUTPUT_RETRY_ATTEMPTS,
+                                )
+                            raise
             except Exception as exc:  # noqa: BLE001
                 last_exception = exc
                 logger.warning(
@@ -302,7 +544,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         message_text: str,
     ) -> FollowUpTicket | None:
         """Queue a follow-up message for the next tool result."""
-        if self.done():
+        if self.done() or self._is_stop_requested():
             return None
         text = (message_text or "").strip()
         if not text:
@@ -331,12 +573,8 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         follow_up_lines = "\n".join(
             f"{idx}. {ticket.text}" for idx, ticket in enumerate(follow_ups, start=1)
         )
-        return (
-            "\n\n[SYSTEM NOTICE] User sent follow-up messages while tool execution "
-            "was in progress. Prioritize these follow-up instructions in your next "
-            "actions. In your very next action, briefly acknowledge to the user "
-            "that their follow-up message(s) were received before continuing.\n"
-            f"{follow_up_lines}"
+        return self.FOLLOW_UP_NOTICE_TEMPLATE.format(
+            follow_up_lines=follow_up_lines,
         )
 
     def _merge_follow_up_notice(self, content: str) -> str:
@@ -345,6 +583,35 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             return content
         return f"{content}{notice}"
 
+    def _track_tool_call_streak(self, tool_name: str) -> int:
+        if tool_name == self._last_tool_name:
+            self._same_tool_streak += 1
+        else:
+            self._last_tool_name = tool_name
+            self._same_tool_streak = 1
+        return self._same_tool_streak
+
+    def _build_repeated_tool_call_guidance(self, tool_name: str, streak: int) -> str:
+        if streak < self.REPEATED_TOOL_NOTICE_L1_THRESHOLD:
+            return ""
+
+        if streak >= self.REPEATED_TOOL_NOTICE_L3_THRESHOLD:
+            return self.REPEATED_TOOL_NOTICE_L3_TEMPLATE.format(
+                tool_name=tool_name,
+                streak=streak,
+            )
+
+        if streak >= self.REPEATED_TOOL_NOTICE_L2_THRESHOLD:
+            return self.REPEATED_TOOL_NOTICE_L2_TEMPLATE.format(
+                tool_name=tool_name,
+                streak=streak,
+            )
+
+        return self.REPEATED_TOOL_NOTICE_L1_TEMPLATE.format(
+            tool_name=tool_name,
+            streak=streak,
+        )
+
     @override
     async def step(self):
         """Process a single step of the agent.
@@ -398,10 +665,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                             ),
                         ),
                     )
-                if self._stop_requested:
+                if self._is_stop_requested():
                     llm_resp_result = LLMResponse(
                         role="assistant",
-                        completion_text="[SYSTEM: User actively interrupted the response generation. Partial output before interruption is preserved.]",
+                        completion_text=self.USER_INTERRUPTION_MESSAGE,
                         reasoning_content=llm_response.reasoning_content,
                         reasoning_signature=llm_response.reasoning_signature,
                     )
@@ -417,49 +684,13 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             break  # got final response
 
         if not llm_resp_result:
-            if self._stop_requested:
+            if self._is_stop_requested():
                 llm_resp_result = LLMResponse(role="assistant", completion_text="")
             else:
                 return
 
-        if self._stop_requested:
-            logger.info("Agent execution was requested to stop by user.")
-            llm_resp = llm_resp_result
-            if llm_resp.role != "assistant":
-                llm_resp = LLMResponse(
-                    role="assistant",
-                    completion_text="[SYSTEM: User actively interrupted the response generation. Partial output before interruption is preserved.]",
-                )
-            self.final_llm_resp = llm_resp
-            self._aborted = True
-            self._transition_state(AgentState.DONE)
-            self.stats.end_time = time.time()
-
-            parts = []
-            if llm_resp.reasoning_content or llm_resp.reasoning_signature:
-                parts.append(
-                    ThinkPart(
-                        think=llm_resp.reasoning_content,
-                        encrypted=llm_resp.reasoning_signature,
-                    )
-                )
-            if llm_resp.completion_text:
-                parts.append(TextPart(text=llm_resp.completion_text))
-            if parts:
-                self.run_context.messages.append(
-                    Message(role="assistant", content=parts)
-                )
-
-            try:
-                await self.agent_hooks.on_agent_done(self.run_context, llm_resp)
-            except Exception as e:
-                logger.error(f"Error in on_agent_done hook: {e}", exc_info=True)
-
-            yield AgentResponse(
-                type="aborted",
-                data=AgentResponseData(chain=MessageChain(type="aborted")),
-            )
-            self._resolve_unconsumed_follow_ups()
+        if self._is_stop_requested():
+            yield await self._finalize_aborted_step(llm_resp_result)
             return
 
         # 处理 LLM 响应
@@ -484,34 +715,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             return
 
         if not llm_resp.tools_call_name:
-            # 如果没有工具调用，转换到完成状态
-            self.final_llm_resp = llm_resp
-            self._transition_state(AgentState.DONE)
-            self.stats.end_time = time.time()
-
-            # record the final assistant message
-            parts = []
-            if llm_resp.reasoning_content or llm_resp.reasoning_signature:
-                parts.append(
-                    ThinkPart(
-                        think=llm_resp.reasoning_content,
-                        encrypted=llm_resp.reasoning_signature,
-                    )
-                )
-            if llm_resp.completion_text:
-                parts.append(TextPart(text=llm_resp.completion_text))
-            if len(parts) == 0:
-                logger.warning(
-                    "LLM returned empty assistant message with no tool calls."
-                )
-            self.run_context.messages.append(Message(role="assistant", content=parts))
-
-            # call the on_agent_done hook
-            try:
-                await self.agent_hooks.on_agent_done(self.run_context, llm_resp)
-            except Exception as e:
-                logger.error(f"Error in on_agent_done hook: {e}", exc_info=True)
-            self._resolve_unconsumed_follow_ups()
+            await self._complete_with_assistant_response(llm_resp)
 
         # 返回 LLM 结果
         if llm_resp.result_chain:
@@ -531,30 +735,52 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         if llm_resp.tools_call_name:
             if self.tool_schema_mode == "skills_like":
                 llm_resp, _ = await self._resolve_tool_exec(llm_resp)
+                if not llm_resp.tools_call_name:
+                    logger.warning(
+                        "skills_like tool re-query returned no tool calls; fallback to assistant response."
+                    )
+                    if llm_resp.result_chain:
+                        yield AgentResponse(
+                            type="llm_result",
+                            data=AgentResponseData(chain=llm_resp.result_chain),
+                        )
+                    elif llm_resp.completion_text:
+                        yield AgentResponse(
+                            type="llm_result",
+                            data=AgentResponseData(
+                                chain=MessageChain().message(llm_resp.completion_text),
+                            ),
+                        )
+                    await self._complete_with_assistant_response(llm_resp)
+                    return
 
             tool_call_result_blocks = []
             cached_images = []  # Collect cached images for LLM visibility
-            async for result in self._handle_function_tools(self.req, llm_resp):
-                if result.kind == "tool_call_result_blocks":
-                    if result.tool_call_result_blocks is not None:
-                        tool_call_result_blocks = result.tool_call_result_blocks
-                elif result.kind == "cached_image":
-                    if result.cached_image is not None:
-                        # Collect cached image info
-                        cached_images.append(result.cached_image)
-                elif result.kind == "message_chain":
-                    chain = result.message_chain
-                    if chain is None or chain.type is None:
-                        # should not happen
-                        continue
-                    if chain.type == "tool_direct_result":
-                        ar_type = "tool_call_result"
-                    else:
-                        ar_type = chain.type
-                    yield AgentResponse(
-                        type=ar_type,
-                        data=AgentResponseData(chain=chain),
-                    )
+            try:
+                async for result in self._handle_function_tools(self.req, llm_resp):
+                    if result.kind == "tool_call_result_blocks":
+                        if result.tool_call_result_blocks is not None:
+                            tool_call_result_blocks = result.tool_call_result_blocks
+                    elif result.kind == "cached_image":
+                        if result.cached_image is not None:
+                            # Collect cached image info
+                            cached_images.append(result.cached_image)
+                    elif result.kind == "message_chain":
+                        chain = result.message_chain
+                        if chain is None or chain.type is None:
+                            # should not happen
+                            continue
+                        if chain.type == "tool_direct_result":
+                            ar_type = "tool_call_result"
+                        else:
+                            ar_type = chain.type
+                        yield AgentResponse(
+                            type=ar_type,
+                            data=AgentResponseData(chain=chain),
+                        )
+            except _ToolExecutionInterrupted:
+                yield await self._finalize_aborted_step(llm_resp)
+                return
 
             # 将结果添加到上下文中
             parts = []
@@ -640,7 +866,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             self.run_context.messages.append(
                 Message(
                     role="user",
-                    content="工具调用次数已达到上限，请停止使用工具，并根据已经收集到的信息，对你的任务和发现进行总结，然后直接回复用户。",
+                    content=self.MAX_STEPS_REACHED_PROMPT,
                 )
             )
             # 再执行最后一步
@@ -671,6 +897,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             llm_response.tools_call_args,
             llm_response.tools_call_ids,
         ):
+            tool_call_streak = self._track_tool_call_streak(func_tool_name)
             yield _HandleFunctionToolsResult.from_message_chain(
                 MessageChain(
                     type="tool_call",
@@ -754,73 +981,82 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                 )
 
                 _final_resp: CallToolResult | None = None
-                async for resp in executor:  # type: ignore
+                async for resp in self._iter_tool_executor_results(executor):  # type: ignore
                     if isinstance(resp, CallToolResult):
                         res = resp
                         _final_resp = resp
-                        if isinstance(res.content[0], TextContent):
+                        if not res.content:
                             _append_tool_call_result(
                                 func_tool_id,
-                                res.content[0].text,
+                                "The tool returned no content.",
                             )
-                        elif isinstance(res.content[0], ImageContent):
-                            # Cache the image instead of sending directly
-                            cached_img = tool_image_cache.save_image(
-                                base64_data=res.content[0].data,
-                                tool_call_id=func_tool_id,
-                                tool_name=func_tool_name,
-                                index=0,
-                                mime_type=res.content[0].mimeType or "image/png",
-                            )
-                            _append_tool_call_result(
-                                func_tool_id,
-                                (
+                            continue
+
+                        result_parts: list[str] = []
+                        for index, content_item in enumerate(res.content):
+                            if isinstance(content_item, TextContent):
+                                result_parts.append(content_item.text)
+                            elif isinstance(content_item, ImageContent):
+                                # Cache the image instead of sending directly
+                                cached_img = tool_image_cache.save_image(
+                                    base64_data=content_item.data,
+                                    tool_call_id=func_tool_id,
+                                    tool_name=func_tool_name,
+                                    index=index,
+                                    mime_type=content_item.mimeType or "image/png",
+                                )
+                                result_parts.append(
                                     f"Image returned and cached at path='{cached_img.file_path}'. "
                                     f"Review the image below. Use send_message_to_user to send it to the user if satisfied, "
                                     f"with type='image' and path='{cached_img.file_path}'."
-                                ),
-                            )
-                            # Yield image info for LLM visibility (will be handled in step())
-                            yield _HandleFunctionToolsResult.from_cached_image(
-                                cached_img
-                            )
-                        elif isinstance(res.content[0], EmbeddedResource):
-                            resource = res.content[0].resource
-                            if isinstance(resource, TextResourceContents):
-                                _append_tool_call_result(
-                                    func_tool_id,
-                                    resource.text,
                                 )
-                            elif (
-                                isinstance(resource, BlobResourceContents)
-                                and resource.mimeType
-                                and resource.mimeType.startswith("image/")
-                            ):
-                                # Cache the image instead of sending directly
-                                cached_img = tool_image_cache.save_image(
-                                    base64_data=resource.blob,
-                                    tool_call_id=func_tool_id,
-                                    tool_name=func_tool_name,
-                                    index=0,
-                                    mime_type=resource.mimeType,
-                                )
-                                _append_tool_call_result(
-                                    func_tool_id,
-                                    (
-                                        f"Image returned and cached at path='{cached_img.file_path}'. "
-                                        f"Review the image below. Use send_message_to_user to send it to the user if satisfied, "
-                                        f"with type='image' and path='{cached_img.file_path}'."
-                                    ),
-                                )
-                                # Yield image info for LLM visibility
+                                # Yield image info for LLM visibility (will be handled in step())
                                 yield _HandleFunctionToolsResult.from_cached_image(
                                     cached_img
                                 )
-                            else:
-                                _append_tool_call_result(
-                                    func_tool_id,
-                                    "The tool has returned a data type that is not supported.",
-                                )
+                            elif isinstance(content_item, EmbeddedResource):
+                                resource = content_item.resource
+                                if isinstance(resource, TextResourceContents):
+                                    result_parts.append(resource.text)
+                                elif (
+                                    isinstance(resource, BlobResourceContents)
+                                    and resource.mimeType
+                                    and resource.mimeType.startswith("image/")
+                                ):
+                                    # Cache the image instead of sending directly
+                                    cached_img = tool_image_cache.save_image(
+                                        base64_data=resource.blob,
+                                        tool_call_id=func_tool_id,
+                                        tool_name=func_tool_name,
+                                        index=index,
+                                        mime_type=resource.mimeType,
+                                    )
+                                    result_parts.append(
+                                        f"Image returned and cached at path='{cached_img.file_path}'. "
+                                        f"Review the image below. Use send_message_to_user to send it to the user if satisfied, "
+                                        f"with type='image' and path='{cached_img.file_path}'."
+                                    )
+                                    # Yield image info for LLM visibility
+                                    yield _HandleFunctionToolsResult.from_cached_image(
+                                        cached_img
+                                    )
+                                else:
+                                    result_parts.append(
+                                        "The tool has returned a data type that is not supported."
+                                    )
+                        if result_parts:
+                            inline_result = "\n\n".join(result_parts)
+                            inline_result = await self._materialize_large_tool_result(
+                                tool_call_id=func_tool_id,
+                                content=inline_result,
+                            )
+                            _append_tool_call_result(
+                                func_tool_id,
+                                inline_result
+                                + self._build_repeated_tool_call_guidance(
+                                    func_tool_name, tool_call_streak
+                                ),
+                            )
 
                     elif resp is None:
                         # Tool 直接请求发送消息给用户
@@ -833,7 +1069,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                         self.stats.end_time = time.time()
                         _append_tool_call_result(
                             func_tool_id,
-                            "The tool has no return value, or has sent the result directly to the user.",
+                            "The tool has no return value, or has sent the result directly to the user."
+                            + self._build_repeated_tool_call_guidance(
+                                func_tool_name, tool_call_streak
+                            ),
                         )
                     else:
                         # 不应该出现其他类型
@@ -842,7 +1081,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                         )
                         _append_tool_call_result(
                             func_tool_id,
-                            "*The tool has returned an unsupported type. Please tell the user to check the definition and implementation of this tool.*",
+                            "*The tool has returned an unsupported type. Please tell the user to check the definition and implementation of this tool.*"
+                            + self._build_repeated_tool_call_guidance(
+                                func_tool_name, tool_call_streak
+                            ),
                         )
 
                 try:
@@ -855,10 +1097,15 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                 except Exception as e:
                     logger.error(f"Error in on_tool_end hook: {e}", exc_info=True)
             except Exception as e:
+                if isinstance(e, _ToolExecutionInterrupted):
+                    raise
                 logger.warning(traceback.format_exc())
                 _append_tool_call_result(
                     func_tool_id,
-                    f"error: {e!s}",
+                    f"error: {e!s}"
+                    + self._build_repeated_tool_call_guidance(
+                        func_tool_name, tool_call_streak
+                    ),
                 )
 
         # yield the last tool call result
@@ -887,7 +1134,9 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             )
 
     def _build_tool_requery_context(
-        self, tool_names: list[str]
+        self,
+        tool_names: list[str],
+        extra_instruction: str | None = None,
     ) -> list[dict[str, T.Any]]:
         """Build contexts for re-querying LLM with param-only tool schemas."""
         contexts: list[dict[str, T.Any]] = []
@@ -896,12 +1145,11 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                 contexts.append(msg.model_dump())  # type: ignore[call-arg]
             elif isinstance(msg, dict):
                 contexts.append(copy.deepcopy(msg))
-        instruction = (
-            "You have decided to call tool(s): "
-            + ", ".join(tool_names)
-            + ". Now call the tool(s) with required arguments using the tool schema, "
-            "and follow the existing tool-use rules."
+        instruction = self.SKILLS_LIKE_REQUERY_INSTRUCTION_TEMPLATE.format(
+            tool_names=", ".join(tool_names)
         )
+        if extra_instruction:
+            instruction = f"{instruction}\n{extra_instruction}"
         if contexts and contexts[0].get("role") == "system":
             content = contexts[0].get("content") or ""
             contexts[0]["content"] = f"{content}\n{instruction}"
@@ -909,6 +1157,11 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
             contexts.insert(0, {"role": "system", "content": instruction})
         return contexts
 
+    @staticmethod
+    def _has_meaningful_assistant_reply(llm_resp: LLMResponse) -> bool:
+        text = (llm_resp.completion_text or "").strip()
+        return bool(text)
+
     def _build_tool_subset(self, tool_set: ToolSet, tool_names: list[str]) -> ToolSet:
         """Build a subset of tools from the given tool set based on tool names."""
         subset = ToolSet()
@@ -945,10 +1198,39 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
                     func_tool=param_subset,
                     model=self.req.model,
                     session_id=self.req.session_id,
+                    extra_user_content_parts=self.req.extra_user_content_parts,
+                    tool_choice="required",
+                    abort_signal=self._abort_signal,
                 )
                 if requery_resp:
                     llm_resp = requery_resp
 
+                # If the re-query still returns no tool calls, and also does not have a meaningful assistant reply,
+                # we consider it as a failure of the LLM to follow the tool-use instruction,
+                # and we will retry once with a stronger instruction that explicitly requires the LLM to either call the tool or give an explanation.
+                if (
+                    not llm_resp.tools_call_name
+                    and not self._has_meaningful_assistant_reply(llm_resp)
+                ):
+                    logger.warning(
+                        "skills_like tool re-query returned no tool calls and no explanation; retrying with stronger instruction."
+                    )
+                    repair_contexts = self._build_tool_requery_context(
+                        tool_names,
+                        extra_instruction=self.SKILLS_LIKE_REQUERY_REPAIR_INSTRUCTION,
+                    )
+                    repair_resp = await self.provider.text_chat(
+                        contexts=repair_contexts,
+                        func_tool=param_subset,
+                        model=self.req.model,
+                        session_id=self.req.session_id,
+                        extra_user_content_parts=self.req.extra_user_content_parts,
+                        tool_choice="required",
+                        abort_signal=self._abort_signal,
+                    )
+                    if repair_resp:
+                        llm_resp = repair_resp
+
         return llm_resp, subset
 
     def done(self) -> bool:
@@ -956,10 +1238,102 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         return self._state in (AgentState.DONE, AgentState.ERROR)
 
     def request_stop(self) -> None:
-        self._stop_requested = True
+        self._abort_signal.set()
+
+    def _is_stop_requested(self) -> bool:
+        return self._abort_signal.is_set()
 
     def was_aborted(self) -> bool:
         return self._aborted
 
     def get_final_llm_resp(self) -> LLMResponse | None:
         return self.final_llm_resp
+
+    async def _finalize_aborted_step(
+        self,
+        llm_resp: LLMResponse | None = None,
+    ) -> AgentResponse:
+        logger.info("Agent execution was requested to stop by user.")
+        if llm_resp is None:
+            llm_resp = LLMResponse(role="assistant", completion_text="")
+        if llm_resp.role != "assistant":
+            llm_resp = LLMResponse(
+                role="assistant",
+                completion_text=self.USER_INTERRUPTION_MESSAGE,
+            )
+        self.final_llm_resp = llm_resp
+        self._aborted = True
+        self._transition_state(AgentState.DONE)
+        self.stats.end_time = time.time()
+
+        parts = []
+        if llm_resp.reasoning_content or llm_resp.reasoning_signature:
+            parts.append(
+                ThinkPart(
+                    think=llm_resp.reasoning_content,
+                    encrypted=llm_resp.reasoning_signature,
+                )
+            )
+        if llm_resp.completion_text:
+            parts.append(TextPart(text=llm_resp.completion_text))
+        if parts:
+            self.run_context.messages.append(Message(role="assistant", content=parts))
+
+        try:
+            await self.agent_hooks.on_agent_done(self.run_context, llm_resp)
+        except Exception as e:
+            logger.error(f"Error in on_agent_done hook: {e}", exc_info=True)
+
+        self._resolve_unconsumed_follow_ups()
+        return AgentResponse(
+            type="aborted",
+            data=AgentResponseData(chain=MessageChain(type="aborted")),
+        )
+
+    async def _close_executor(self, executor: T.Any) -> None:
+        close_executor = getattr(executor, "aclose", None)
+        if close_executor is None:
+            return
+        with suppress(asyncio.CancelledError, RuntimeError, StopAsyncIteration):
+            await close_executor()
+
+    async def _iter_tool_executor_results(
+        self,
+        executor: AsyncIterator[ToolExecutorResultT],
+    ) -> T.AsyncGenerator[ToolExecutorResultT, None]:
+        while True:
+            if self._is_stop_requested():
+                await self._close_executor(executor)
+                raise _ToolExecutionInterrupted(
+                    "Tool execution interrupted before reading the next tool result."
+                )
+
+            next_result_task = asyncio.create_task(anext(executor))
+            abort_task = asyncio.create_task(self._abort_signal.wait())
+            try:
+                done, _ = await asyncio.wait(
+                    {next_result_task, abort_task},
+                    return_when=asyncio.FIRST_COMPLETED,
+                )
+
+                if abort_task in done:
+                    if not next_result_task.done():
+                        next_result_task.cancel()
+                        with suppress(asyncio.CancelledError, StopAsyncIteration):
+                            await next_result_task
+
+                    await self._close_executor(executor)
+
+                    raise _ToolExecutionInterrupted(
+                        "Tool execution interrupted by a stop request."
+                    )
+
+                try:
+                    yield next_result_task.result()
+                except StopAsyncIteration:
+                    return
+            finally:
+                if not abort_task.done():
+                    abort_task.cancel()
+                    with suppress(asyncio.CancelledError):
+                        await abort_task

astrbot/core/agent/tool.py

@@ -89,11 +89,21 @@ class ToolSet:
         return len(self.tools) == 0
 
     def add_tool(self, tool: FunctionTool) -> None:
-        """Add a tool to the set."""
-        # 检查是否已存在同名工具
+        """Add a tool to the set.
+
+        If a tool with the same name already exists:
+        - Prefer the one that is active (active=True)
+        - If both have the same active state, use the new one (overwrite)
+        """
         for i, existing_tool in enumerate(self.tools):
             if existing_tool.name == tool.name:
-                self.tools[i] = tool
+                # Use getattr with a default of True for compatibility with tools
+                # that may not define an `active` attribute (e.g., mocks).
+                existing_active = bool(getattr(existing_tool, "active", True))
+                new_active = bool(getattr(tool, "active", True))
+                # Overwrite if new tool is active, or if existing tool is not active
+                if new_active or not existing_active:
+                    self.tools[i] = tool
                 return
         self.tools.append(tool)
 
@@ -293,8 +303,15 @@ class ToolSet:
                 if properties:
                     result["properties"] = properties
 
-            if "items" in schema:
-                result["items"] = convert_schema(schema["items"])
+            if target_type == "array":
+                items_schema = schema.get("items")
+                if isinstance(items_schema, dict):
+                    result["items"] = convert_schema(items_schema)
+                else:
+                    # Gemini requires array schemas to include an `items` schema.
+                    # JSON Schema allows omitting it, so fall back to a permissive
+                    # string item schema instead of emitting an invalid declaration.
+                    result["items"] = {"type": "string"}
 
             return result
 

astrbot/core/astr_agent_hooks.py

@@ -59,7 +59,13 @@ class MainAgentHooks(BaseAgentRunHooks[AstrAgentContext]):
         platform_name = run_context.context.event.get_platform_name()
         if (
             platform_name == "webchat"
-            and tool.name in ["web_search_tavily", "web_search_bocha"]
+            and tool.name
+            in [
+                "web_search_baidu",
+                "web_search_tavily",
+                "web_search_bocha",
+                "web_search_brave",
+            ]
             and len(run_context.messages) > 0
             and tool_result
             and len(tool_result.content)

astrbot/core/astr_agent_run_util.py

@@ -165,8 +165,13 @@ async def run_agent(
                     # 对于其他情况，暂时先不处理
                     continue
                 elif resp.type == "tool_call":
-                    if agent_runner.streaming:
-                        # 用来标记流式响应需要分节
+                    if agent_runner.streaming and show_tool_use:
+                        # 向下游平台发送 "break" 分段信号（空 MessageChain，不携带数据）。
+                        # 平台适配器收到后会关闭当前流式消息，并在后续文本到来时创建新消息。
+                        # 仅在 show_tool_use 为 True 时才发送：此时紧接着会通过
+                        # astr_event.send() 独立发送工具状态消息（如"🔨 调用工具: xxx"），
+                        # 需要分段才能保证消息顺序正确。
+                        # 若 show_tool_use 为 False，不会有独立消息插入，无需分段。
                         yield MessageChain(chain=[], type="break")
 
                     tool_info = _extract_chain_json_data(resp.data["chain"])

astrbot/core/astr_agent_tool_exec.py

@@ -19,13 +19,6 @@ from astrbot.core.agent.tool_executor import BaseFunctionToolExecutor
 from astrbot.core.astr_agent_context import AstrAgentContext
 from astrbot.core.astr_main_agent_resources import (
     BACKGROUND_TASK_RESULT_WOKE_SYSTEM_PROMPT,
-    EXECUTE_SHELL_TOOL,
-    FILE_DOWNLOAD_TOOL,
-    FILE_UPLOAD_TOOL,
-    LOCAL_EXECUTE_SHELL_TOOL,
-    LOCAL_PYTHON_TOOL,
-    PYTHON_TOOL,
-    SEND_MESSAGE_TO_USER_TOOL,
 )
 from astrbot.core.cron.events import CronMessageEvent
 from astrbot.core.message.components import Image
@@ -37,6 +30,18 @@ from astrbot.core.message.message_event_result import (
 from astrbot.core.platform.message_session import MessageSession
 from astrbot.core.provider.entites import ProviderRequest
 from astrbot.core.provider.register import llm_tools
+from astrbot.core.tools.computer_tools import (
+    ExecuteShellTool,
+    FileDownloadTool,
+    FileEditTool,
+    FileReadTool,
+    FileUploadTool,
+    FileWriteTool,
+    GrepTool,
+    LocalPythonTool,
+    PythonTool,
+)
+from astrbot.core.tools.message_tools import SendMessageToUserTool
 from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.history_saver import persist_agent_history
 from astrbot.core.utils.image_ref_utils import is_supported_image_ref
@@ -177,18 +182,44 @@ class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
             return
 
     @classmethod
-    def _get_runtime_computer_tools(cls, runtime: str) -> dict[str, FunctionTool]:
+    def _get_runtime_computer_tools(
+        cls,
+        runtime: str,
+        tool_mgr,
+    ) -> dict[str, FunctionTool]:
         if runtime == "sandbox":
+            shell_tool = tool_mgr.get_builtin_tool(ExecuteShellTool)
+            python_tool = tool_mgr.get_builtin_tool(PythonTool)
+            upload_tool = tool_mgr.get_builtin_tool(FileUploadTool)
+            download_tool = tool_mgr.get_builtin_tool(FileDownloadTool)
+            read_tool = tool_mgr.get_builtin_tool(FileReadTool)
+            write_tool = tool_mgr.get_builtin_tool(FileWriteTool)
+            edit_tool = tool_mgr.get_builtin_tool(FileEditTool)
+            grep_tool = tool_mgr.get_builtin_tool(GrepTool)
             return {
-                EXECUTE_SHELL_TOOL.name: EXECUTE_SHELL_TOOL,
-                PYTHON_TOOL.name: PYTHON_TOOL,
-                FILE_UPLOAD_TOOL.name: FILE_UPLOAD_TOOL,
-                FILE_DOWNLOAD_TOOL.name: FILE_DOWNLOAD_TOOL,
+                shell_tool.name: shell_tool,
+                python_tool.name: python_tool,
+                upload_tool.name: upload_tool,
+                download_tool.name: download_tool,
+                read_tool.name: read_tool,
+                write_tool.name: write_tool,
+                edit_tool.name: edit_tool,
+                grep_tool.name: grep_tool,
             }
         if runtime == "local":
+            shell_tool = tool_mgr.get_builtin_tool(ExecuteShellTool)
+            python_tool = tool_mgr.get_builtin_tool(LocalPythonTool)
+            read_tool = tool_mgr.get_builtin_tool(FileReadTool)
+            write_tool = tool_mgr.get_builtin_tool(FileWriteTool)
+            edit_tool = tool_mgr.get_builtin_tool(FileEditTool)
+            grep_tool = tool_mgr.get_builtin_tool(GrepTool)
             return {
-                LOCAL_EXECUTE_SHELL_TOOL.name: LOCAL_EXECUTE_SHELL_TOOL,
-                LOCAL_PYTHON_TOOL.name: LOCAL_PYTHON_TOOL,
+                shell_tool.name: shell_tool,
+                python_tool.name: python_tool,
+                read_tool.name: read_tool,
+                write_tool.name: write_tool,
+                edit_tool.name: edit_tool,
+                grep_tool.name: grep_tool,
             }
         return {}
 
@@ -203,7 +234,15 @@ class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
         cfg = ctx.get_config(umo=event.unified_msg_origin)
         provider_settings = cfg.get("provider_settings", {})
         runtime = str(provider_settings.get("computer_use_runtime", "local"))
-        runtime_computer_tools = cls._get_runtime_computer_tools(runtime)
+        tool_mgr = (
+            ctx.get_llm_tool_manager()
+            if hasattr(ctx, "get_llm_tool_manager")
+            else llm_tools
+        )
+        runtime_computer_tools = cls._get_runtime_computer_tools(
+            runtime,
+            tool_mgr,
+        )
 
         # Keep persona semantics aligned with the main agent: tools=None means
         # "all tools", including runtime computer-use tools.
@@ -303,6 +342,7 @@ class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
             tools=toolset,
             contexts=contexts,
             max_steps=agent_max_step,
+            tool_call_timeout=run_context.tool_call_timeout,
             stream=stream,
         )
         yield mcp.types.CallToolResult(
@@ -481,7 +521,7 @@ class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
         )
         cron_event.role = event.role
         config = MainAgentBuildConfig(
-            tool_call_timeout=3600,
+            tool_call_timeout=run_context.tool_call_timeout,
             streaming_response=ctx.get_config()
             .get("provider_settings", {})
             .get("stream", False),
@@ -514,7 +554,9 @@ class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
         )
         if not req.func_tool:
             req.func_tool = ToolSet()
-        req.func_tool.add_tool(SEND_MESSAGE_TO_USER_TOOL)
+        req.func_tool.add_tool(
+            ctx.get_llm_tool_manager().get_builtin_tool(SendMessageToUserTool)
+        )
 
         result = await build_main_agent(
             event=cron_event, plugin_context=ctx, config=config, req=req

astrbot/core/astr_main_agent.py

@@ -9,6 +9,7 @@ import platform
 import zoneinfo
 from collections.abc import Coroutine
 from dataclasses import dataclass, field
+from pathlib import Path
 
 from astrbot.core import logger
 from astrbot.core.agent.handoff import HandoffTool
@@ -20,38 +21,15 @@ from astrbot.core.astr_agent_hooks import MAIN_AGENT_HOOKS
 from astrbot.core.astr_agent_run_util import AgentRunner
 from astrbot.core.astr_agent_tool_exec import FunctionToolExecutor
 from astrbot.core.astr_main_agent_resources import (
-    ANNOTATE_EXECUTION_TOOL,
-    BROWSER_BATCH_EXEC_TOOL,
-    BROWSER_EXEC_TOOL,
     CHATUI_SPECIAL_DEFAULT_PERSONA_PROMPT,
-    CREATE_SKILL_CANDIDATE_TOOL,
-    CREATE_SKILL_PAYLOAD_TOOL,
-    EVALUATE_SKILL_CANDIDATE_TOOL,
-    EXECUTE_SHELL_TOOL,
-    FILE_DOWNLOAD_TOOL,
-    FILE_UPLOAD_TOOL,
-    GET_EXECUTION_HISTORY_TOOL,
-    GET_SKILL_PAYLOAD_TOOL,
-    KNOWLEDGE_BASE_QUERY_TOOL,
-    LIST_SKILL_CANDIDATES_TOOL,
-    LIST_SKILL_RELEASES_TOOL,
     LIVE_MODE_SYSTEM_PROMPT,
     LLM_SAFETY_MODE_SYSTEM_PROMPT,
-    LOCAL_EXECUTE_SHELL_TOOL,
-    LOCAL_PYTHON_TOOL,
-    PROMOTE_SKILL_CANDIDATE_TOOL,
-    PYTHON_TOOL,
-    ROLLBACK_SKILL_RELEASE_TOOL,
-    RUN_BROWSER_SKILL_TOOL,
     SANDBOX_MODE_PROMPT,
-    SEND_MESSAGE_TO_USER_TOOL,
-    SYNC_SKILL_RELEASE_TOOL,
     TOOL_CALL_PROMPT,
     TOOL_CALL_PROMPT_SKILLS_LIKE_MODE,
-    retrieve_knowledge_base,
 )
 from astrbot.core.conversation_mgr import Conversation
-from astrbot.core.message.components import File, Image, Reply
+from astrbot.core.message.components import File, Image, Record, Reply
 from astrbot.core.persona_error_reply import (
     extract_persona_custom_error_message_from_persona,
     set_persona_custom_error_message_on_event,
@@ -59,16 +37,61 @@ from astrbot.core.persona_error_reply import (
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.provider import Provider
 from astrbot.core.provider.entities import ProviderRequest
+from astrbot.core.provider.register import llm_tools
 from astrbot.core.skills.skill_manager import SkillManager, build_skills_prompt
 from astrbot.core.star.context import Context
 from astrbot.core.star.star_handler import star_map
-from astrbot.core.tools.cron_tools import (
-    CREATE_CRON_JOB_TOOL,
-    DELETE_CRON_JOB_TOOL,
-    LIST_CRON_JOBS_TOOL,
+from astrbot.core.tools.computer_tools import (
+    AnnotateExecutionTool,
+    BrowserBatchExecTool,
+    BrowserExecTool,
+    CreateSkillCandidateTool,
+    CreateSkillPayloadTool,
+    EvaluateSkillCandidateTool,
+    ExecuteShellTool,
+    FileDownloadTool,
+    FileEditTool,
+    FileReadTool,
+    FileUploadTool,
+    FileWriteTool,
+    GetExecutionHistoryTool,
+    GetSkillPayloadTool,
+    GrepTool,
+    ListSkillCandidatesTool,
+    ListSkillReleasesTool,
+    LocalPythonTool,
+    PromoteSkillCandidateTool,
+    PythonTool,
+    RollbackSkillReleaseTool,
+    RunBrowserSkillTool,
+    SyncSkillReleaseTool,
+    normalize_umo_for_workspace,
+)
+from astrbot.core.tools.cron_tools import FutureTaskTool
+from astrbot.core.tools.knowledge_base_tools import (
+    KnowledgeBaseQueryTool,
+    retrieve_knowledge_base,
+)
+from astrbot.core.tools.message_tools import SendMessageToUserTool
+from astrbot.core.tools.web_search_tools import (
+    BaiduWebSearchTool,
+    BochaWebSearchTool,
+    BraveWebSearchTool,
+    TavilyExtractWebPageTool,
+    TavilyWebSearchTool,
+    normalize_legacy_web_search_config,
+)
+from astrbot.core.utils.astrbot_path import (
+    get_astrbot_system_tmp_path,
+    get_astrbot_workspaces_path,
 )
 from astrbot.core.utils.file_extract import extract_file_moonshotai
 from astrbot.core.utils.llm_metadata import LLM_METADATAS
+from astrbot.core.utils.media_utils import (
+    IMAGE_COMPRESS_DEFAULT_MAX_SIZE,
+    IMAGE_COMPRESS_DEFAULT_QUALITY,
+    compress_image,
+)
 from astrbot.core.utils.quoted_message.settings import (
     SETTINGS as DEFAULT_QUOTED_MESSAGE_SETTINGS,
 )
@@ -213,7 +236,11 @@ async def _apply_kb(
     else:
         if req.func_tool is None:
             req.func_tool = ToolSet()
-        req.func_tool.add_tool(KNOWLEDGE_BASE_QUERY_TOOL)
+        req.func_tool.add_tool(
+            plugin_context.get_llm_tool_manager().get_builtin_tool(
+                KnowledgeBaseQueryTool
+            )
+        )
 
 
 async def _apply_file_extract(
@@ -275,11 +302,54 @@ def _apply_prompt_prefix(req: ProviderRequest, cfg: dict) -> None:
         req.prompt = f"{prefix}{req.prompt}"
 
 
-def _apply_local_env_tools(req: ProviderRequest) -> None:
+def _get_workspace_path_for_umo(umo: str) -> Path:
+    normalized_umo = normalize_umo_for_workspace(umo)
+    return Path(get_astrbot_workspaces_path()) / normalized_umo
+
+
+def _apply_workspace_extra_prompt(
+    event: AstrMessageEvent,
+    req: ProviderRequest,
+) -> None:
+    extra_prompt_path = _get_workspace_path_for_umo(event.unified_msg_origin) / (
+        "EXTRA_PROMPT.md"
+    )
+    if not extra_prompt_path.is_file():
+        return
+
+    try:
+        extra_prompt = extra_prompt_path.read_text(encoding="utf-8").strip()
+    except Exception as exc:  # noqa: BLE001
+        logger.warning(
+            "Failed to read workspace extra prompt for umo=%s from %s: %s",
+            event.unified_msg_origin,
+            extra_prompt_path,
+            exc,
+        )
+        return
+
+    if not extra_prompt:
+        return
+
+    req.system_prompt = (
+        f"{req.system_prompt or ''}\n"
+        "[Workspace Extra Prompt]\n"
+        "The following instructions are loaded from the current workspace "
+        "`EXTRA_PROMPT.md` file.\n"
+        f"{extra_prompt}\n"
+    )
+
+
+def _apply_local_env_tools(req: ProviderRequest, plugin_context: Context) -> None:
     if req.func_tool is None:
         req.func_tool = ToolSet()
-    req.func_tool.add_tool(LOCAL_EXECUTE_SHELL_TOOL)
-    req.func_tool.add_tool(LOCAL_PYTHON_TOOL)
+    tool_mgr = plugin_context.get_llm_tool_manager()
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(ExecuteShellTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(LocalPythonTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileReadTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileWriteTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileEditTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(GrepTool))
     req.system_prompt = f"{req.system_prompt or ''}\n{_build_local_mode_prompt()}\n"
 
 
@@ -390,14 +460,9 @@ async def _ensure_persona_and_skills(
                 persona_tools = None
                 pid = a.get("persona_id")
                 if pid:
-                    persona_tools = next(
-                        (
-                            p.get("tools")
-                            for p in plugin_context.persona_manager.personas_v3
-                            if p["name"] == pid
-                        ),
-                        None,
-                    )
+                    persona = plugin_context.persona_manager.get_persona_v3_by_id(pid)
+                    if persona is not None:
+                        persona_tools = persona.get("tools")
                 tools = a.get("tools", [])
                 if persona_tools is not None:
                     tools = persona_tools
@@ -478,16 +543,23 @@ async def _request_img_caption(
 
 
 async def _ensure_img_caption(
+    event: AstrMessageEvent,
     req: ProviderRequest,
     cfg: dict,
     plugin_context: Context,
     image_caption_provider: str,
 ) -> None:
     try:
+        compressed_urls = []
+        for url in req.image_urls:
+            compressed_url = await _compress_image_for_provider(url, cfg)
+            compressed_urls.append(compressed_url)
+            if _is_generated_compressed_image_path(url, compressed_url):
+                event.track_temporary_local_file(compressed_url)
         caption = await _request_img_caption(
             image_caption_provider,
             cfg,
-            req.image_urls,
+            compressed_urls,
             plugin_context,
         )
         if caption:
@@ -497,6 +569,9 @@ async def _ensure_img_caption(
             req.image_urls = []
     except Exception as exc:  # noqa: BLE001
         logger.error("处理图片描述失败: %s", exc)
+        req.extra_user_content_parts.append(TextPart(text="[Image Captioning Failed]"))
+    finally:
+        req.image_urls = []
 
 
 def _append_quoted_image_attachment(req: ProviderRequest, image_path: str) -> None:
@@ -505,6 +580,18 @@ def _append_quoted_image_attachment(req: ProviderRequest, image_path: str) -> No
     )
 
 
+def _append_audio_attachment(req: ProviderRequest, audio_path: str) -> None:
+    req.extra_user_content_parts.append(
+        TextPart(text=f"[Audio Attachment: path {audio_path}]")
+    )
+
+
+def _append_quoted_audio_attachment(req: ProviderRequest, audio_path: str) -> None:
+    req.extra_user_content_parts.append(
+        TextPart(text=f"[Audio Attachment in quoted message: path {audio_path}]")
+    )
+
+
 def _get_quoted_message_parser_settings(
     provider_settings: dict[str, object] | None,
 ) -> QuotedMessageParserSettings:
@@ -516,12 +603,64 @@ def _get_quoted_message_parser_settings(
     return DEFAULT_QUOTED_MESSAGE_SETTINGS.with_overrides(overrides)
 
 
+def _get_image_compress_args(
+    provider_settings: dict[str, object] | None,
+) -> tuple[bool, int, int]:
+    if not isinstance(provider_settings, dict):
+        return True, IMAGE_COMPRESS_DEFAULT_MAX_SIZE, IMAGE_COMPRESS_DEFAULT_QUALITY
+
+    enabled = provider_settings.get("image_compress_enabled", True)
+    if not isinstance(enabled, bool):
+        enabled = True
+
+    raw_options = provider_settings.get("image_compress_options", {})
+    options = raw_options if isinstance(raw_options, dict) else {}
+
+    max_size = options.get("max_size", IMAGE_COMPRESS_DEFAULT_MAX_SIZE)
+    if not isinstance(max_size, int):
+        max_size = IMAGE_COMPRESS_DEFAULT_MAX_SIZE
+    max_size = max(max_size, 1)
+
+    quality = options.get("quality", IMAGE_COMPRESS_DEFAULT_QUALITY)
+    if not isinstance(quality, int):
+        quality = IMAGE_COMPRESS_DEFAULT_QUALITY
+    quality = min(max(quality, 1), 100)
+
+    return enabled, max_size, quality
+
+
+async def _compress_image_for_provider(
+    url_or_path: str,
+    provider_settings: dict[str, object] | None,
+) -> str:
+    try:
+        enabled, max_size, quality = _get_image_compress_args(provider_settings)
+        if not enabled:
+            return url_or_path
+        return await compress_image(url_or_path, max_size=max_size, quality=quality)
+    except Exception as exc:  # noqa: BLE001
+        logger.error("Image compression failed: %s", exc)
+        return url_or_path
+
+
+def _is_generated_compressed_image_path(
+    original_path: str,
+    compressed_path: str | None,
+) -> bool:
+    if not compressed_path or compressed_path == original_path:
+        return False
+    if compressed_path.startswith("http") or compressed_path.startswith("data:image"):
+        return False
+    return os.path.exists(compressed_path)
+
+
 async def _process_quote_message(
     event: AstrMessageEvent,
     req: ProviderRequest,
     img_cap_prov_id: str,
     plugin_context: Context,
     quoted_message_settings: QuotedMessageParserSettings = DEFAULT_QUOTED_MESSAGE_SETTINGS,
+    config: MainAgentBuildConfig | None = None,
 ) -> None:
     quote = None
     for comp in event.message_obj.message:
@@ -554,15 +693,24 @@ async def _process_quote_message(
     if image_seg:
         try:
             prov = None
+            path = None
+            compress_path = None
             if img_cap_prov_id:
                 prov = plugin_context.get_provider_by_id(img_cap_prov_id)
             if prov is None:
                 prov = plugin_context.get_using_provider(event.unified_msg_origin)
 
             if prov and isinstance(prov, Provider):
+                path = await image_seg.convert_to_file_path()
+                compress_path = await _compress_image_for_provider(
+                    path,
+                    config.provider_settings if config else None,
+                )
+                if path and _is_generated_compressed_image_path(path, compress_path):
+                    event.track_temporary_local_file(compress_path)
                 llm_resp = await prov.text_chat(
                     prompt="Please describe the image content.",
-                    image_urls=[await image_seg.convert_to_file_path()],
+                    image_urls=[compress_path],
                 )
                 if llm_resp.completion_text:
                     content_parts.append(
@@ -572,6 +720,16 @@ async def _process_quote_message(
                 logger.warning("No provider found for image captioning in quote.")
         except BaseException as exc:
             logger.error("处理引用图片失败: %s", exc)
+        finally:
+            if (
+                compress_path
+                and compress_path != path
+                and os.path.exists(compress_path)
+            ):
+                try:
+                    os.remove(compress_path)
+                except Exception as exc:  # noqa: BLE001
+                    logger.warning("Fail to remove temporary compressed image: %s", exc)
 
     quoted_content = "\n".join(content_parts)
     quoted_text = f"<Quoted Message>\n{quoted_content}\n</Quoted Message>"
@@ -640,6 +798,7 @@ async def _decorate_llm_request(
         img_cap_prov_id: str = cfg.get("default_image_caption_provider_id") or ""
         if img_cap_prov_id and req.image_urls:
             await _ensure_img_caption(
+                event,
                 req,
                 cfg,
                 plugin_context,
@@ -654,12 +813,14 @@ async def _decorate_llm_request(
         img_cap_prov_id,
         plugin_context,
         quoted_message_settings,
+        config,
     )
 
     tz = config.timezone
     if tz is None:
         tz = plugin_context.get_config().get("timezone")
     _append_system_reminders(event, req, cfg, tz)
+    _apply_workspace_extra_prompt(event, req)
 
 
 def _modalities_fix(provider: Provider, req: ProviderRequest) -> None:
@@ -670,12 +831,25 @@ def _modalities_fix(provider: Provider, req: ProviderRequest) -> None:
                 "Provider %s does not support image, using placeholder.", provider
             )
             image_count = len(req.image_urls)
-            placeholder = " ".join(["[图片]"] * image_count)
+            placeholder = " ".join(["[Image]"] * image_count)
             if req.prompt:
                 req.prompt = f"{placeholder} {req.prompt}"
             else:
                 req.prompt = placeholder
             req.image_urls = []
+    if req.audio_urls:
+        provider_cfg = provider.provider_config.get("modalities", ["audio"])
+        if "audio" not in provider_cfg:
+            logger.debug(
+                "Provider %s does not support audio, using placeholder.", provider
+            )
+            audio_count = len(req.audio_urls)
+            placeholder = " ".join(["[Audio]"] * audio_count)
+            if req.prompt:
+                req.prompt = f"{placeholder} {req.prompt}"
+            else:
+                req.prompt = placeholder
+            req.audio_urls = []
     if req.func_tool:
         provider_cfg = provider.provider_config.get("modalities", ["tool_use"])
         if "tool_use" not in provider_cfg:
@@ -698,12 +872,14 @@ def _sanitize_context_by_modalities(
     if not modalities or not isinstance(modalities, list):
         return
     supports_image = bool("image" in modalities)
+    supports_audio = bool("audio" in modalities)
     supports_tool_use = bool("tool_use" in modalities)
-    if supports_image and supports_tool_use:
+    if supports_image and supports_audio and supports_tool_use:
         return
 
     sanitized_contexts: list[dict] = []
     removed_image_blocks = 0
+    removed_audio_blocks = 0
     removed_tool_messages = 0
     removed_tool_calls = 0
 
@@ -725,20 +901,27 @@ def _sanitize_context_by_modalities(
                 new_msg.pop("tool_calls", None)
                 new_msg.pop("tool_call_id", None)
 
-        if not supports_image:
+        if not supports_image or not supports_audio:
             content = new_msg.get("content")
             if isinstance(content, list):
                 filtered_parts: list = []
-                removed_any_image = False
+                removed_any_multimodal = False
                 for part in content:
                     if isinstance(part, dict):
                         part_type = str(part.get("type", "")).lower()
-                        if part_type in {"image_url", "image"}:
-                            removed_any_image = True
+                        if not supports_image and part_type in {"image_url", "image"}:
+                            removed_any_multimodal = True
                             removed_image_blocks += 1
                             continue
+                        if not supports_audio and part_type in {
+                            "audio_url",
+                            "input_audio",
+                        }:
+                            removed_any_multimodal = True
+                            removed_audio_blocks += 1
+                            continue
                     filtered_parts.append(part)
-                if removed_any_image:
+                if removed_any_multimodal:
                     new_msg["content"] = filtered_parts
 
         if role == "assistant":
@@ -752,11 +935,18 @@ def _sanitize_context_by_modalities(
 
         sanitized_contexts.append(new_msg)
 
-    if removed_image_blocks or removed_tool_messages or removed_tool_calls:
+    if (
+        removed_image_blocks
+        or removed_audio_blocks
+        or removed_tool_messages
+        or removed_tool_calls
+    ):
         logger.debug(
             "sanitize_context_by_modalities applied: "
-            "removed_image_blocks=%s, removed_tool_messages=%s, removed_tool_calls=%s",
+            "removed_image_blocks=%s, removed_audio_blocks=%s, "
+            "removed_tool_messages=%s, removed_tool_calls=%s",
             removed_image_blocks,
+            removed_audio_blocks,
             removed_tool_messages,
             removed_tool_calls,
         )
@@ -847,7 +1037,9 @@ def _apply_llm_safety_mode(config: MainAgentBuildConfig, req: ProviderRequest) -
 
 
 def _apply_sandbox_tools(
-    config: MainAgentBuildConfig, req: ProviderRequest, session_id: str
+    config: MainAgentBuildConfig,
+    req: ProviderRequest,
+    session_id: str,
 ) -> None:
     if req.func_tool is None:
         req.func_tool = ToolSet()
@@ -863,10 +1055,15 @@ def _apply_sandbox_tools(
         os.environ["SHIPYARD_ENDPOINT"] = ep
         os.environ["SHIPYARD_ACCESS_TOKEN"] = at
 
-    req.func_tool.add_tool(EXECUTE_SHELL_TOOL)
-    req.func_tool.add_tool(PYTHON_TOOL)
-    req.func_tool.add_tool(FILE_UPLOAD_TOOL)
-    req.func_tool.add_tool(FILE_DOWNLOAD_TOOL)
+    tool_mgr = llm_tools
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(ExecuteShellTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(PythonTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileUploadTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileDownloadTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileReadTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileWriteTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FileEditTool))
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(GrepTool))
     if booter == "shipyard_neo":
         # Neo-specific path rule: filesystem tools operate relative to sandbox
         # workspace root. Do not prepend "/workspace".
@@ -902,32 +1099,59 @@ def _apply_sandbox_tools(
         # Browser tools: only register if profile supports browser
         # (or if capabilities are unknown because sandbox hasn't booted yet)
         if sandbox_capabilities is None or "browser" in sandbox_capabilities:
-            req.func_tool.add_tool(BROWSER_EXEC_TOOL)
-            req.func_tool.add_tool(BROWSER_BATCH_EXEC_TOOL)
-            req.func_tool.add_tool(RUN_BROWSER_SKILL_TOOL)
+            req.func_tool.add_tool(tool_mgr.get_builtin_tool(BrowserExecTool))
+            req.func_tool.add_tool(tool_mgr.get_builtin_tool(BrowserBatchExecTool))
+            req.func_tool.add_tool(tool_mgr.get_builtin_tool(RunBrowserSkillTool))
 
         # Neo-specific tools (always available for shipyard_neo)
-        req.func_tool.add_tool(GET_EXECUTION_HISTORY_TOOL)
-        req.func_tool.add_tool(ANNOTATE_EXECUTION_TOOL)
-        req.func_tool.add_tool(CREATE_SKILL_PAYLOAD_TOOL)
-        req.func_tool.add_tool(GET_SKILL_PAYLOAD_TOOL)
-        req.func_tool.add_tool(CREATE_SKILL_CANDIDATE_TOOL)
-        req.func_tool.add_tool(LIST_SKILL_CANDIDATES_TOOL)
-        req.func_tool.add_tool(EVALUATE_SKILL_CANDIDATE_TOOL)
-        req.func_tool.add_tool(PROMOTE_SKILL_CANDIDATE_TOOL)
-        req.func_tool.add_tool(LIST_SKILL_RELEASES_TOOL)
-        req.func_tool.add_tool(ROLLBACK_SKILL_RELEASE_TOOL)
-        req.func_tool.add_tool(SYNC_SKILL_RELEASE_TOOL)
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(GetExecutionHistoryTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(AnnotateExecutionTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(CreateSkillPayloadTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(GetSkillPayloadTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(CreateSkillCandidateTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(ListSkillCandidatesTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(EvaluateSkillCandidateTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(PromoteSkillCandidateTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(ListSkillReleasesTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(RollbackSkillReleaseTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(SyncSkillReleaseTool))
 
     req.system_prompt = f"{req.system_prompt or ''}\n{SANDBOX_MODE_PROMPT}\n"
 
 
-def _proactive_cron_job_tools(req: ProviderRequest) -> None:
+def _proactive_cron_job_tools(req: ProviderRequest, plugin_context: Context) -> None:
     if req.func_tool is None:
         req.func_tool = ToolSet()
-    req.func_tool.add_tool(CREATE_CRON_JOB_TOOL)
-    req.func_tool.add_tool(DELETE_CRON_JOB_TOOL)
-    req.func_tool.add_tool(LIST_CRON_JOBS_TOOL)
+    tool_mgr = plugin_context.get_llm_tool_manager()
+    req.func_tool.add_tool(tool_mgr.get_builtin_tool(FutureTaskTool))
+
+
+async def _apply_web_search_tools(
+    event: AstrMessageEvent,
+    req: ProviderRequest,
+    plugin_context: Context,
+) -> None:
+    cfg = plugin_context.get_config(umo=event.unified_msg_origin)
+    normalize_legacy_web_search_config(cfg)
+    prov_settings = cfg.get("provider_settings", {})
+
+    if not prov_settings.get("web_search", False):
+        return
+
+    if req.func_tool is None:
+        req.func_tool = ToolSet()
+
+    tool_mgr = plugin_context.get_llm_tool_manager()
+    provider = prov_settings.get("websearch_provider", "tavily")
+    if provider == "tavily":
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(TavilyWebSearchTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(TavilyExtractWebPageTool))
+    elif provider == "bocha":
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(BochaWebSearchTool))
+    elif provider == "brave":
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(BraveWebSearchTool))
+    elif provider == "baidu_ai_search":
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(BaiduWebSearchTool))
 
 
 def _get_compress_provider(
@@ -1018,6 +1242,7 @@ async def build_main_agent(
             req = ProviderRequest()
             req.prompt = ""
             req.image_urls = []
+            req.audio_urls = []
             if sel_model := event.get_extra("selected_model"):
                 req.model = sel_model
             if config.provider_wake_prefix and not event.message_str.startswith(
@@ -1030,11 +1255,21 @@ async def build_main_agent(
             # media files attachments
             for comp in event.message_obj.message:
                 if isinstance(comp, Image):
-                    image_path = await comp.convert_to_file_path()
+                    path = await comp.convert_to_file_path()
+                    image_path = await _compress_image_for_provider(
+                        path,
+                        config.provider_settings,
+                    )
+                    if _is_generated_compressed_image_path(path, image_path):
+                        event.track_temporary_local_file(image_path)
                     req.image_urls.append(image_path)
                     req.extra_user_content_parts.append(
                         TextPart(text=f"[Image Attachment: path {image_path}]")
                     )
+                elif isinstance(comp, Record):
+                    audio_path = await comp.convert_to_file_path()
+                    req.audio_urls.append(audio_path)
+                    _append_audio_attachment(req, audio_path)
                 elif isinstance(comp, File):
                     file_path = await comp.get_file()
                     file_name = comp.name or os.path.basename(file_path)
@@ -1057,9 +1292,19 @@ async def build_main_agent(
                     for reply_comp in comp.chain:
                         if isinstance(reply_comp, Image):
                             has_embedded_image = True
-                            image_path = await reply_comp.convert_to_file_path()
+                            path = await reply_comp.convert_to_file_path()
+                            image_path = await _compress_image_for_provider(
+                                path,
+                                config.provider_settings,
+                            )
+                            if _is_generated_compressed_image_path(path, image_path):
+                                event.track_temporary_local_file(image_path)
                             req.image_urls.append(image_path)
                             _append_quoted_image_attachment(req, image_path)
+                        elif isinstance(reply_comp, Record):
+                            audio_path = await reply_comp.convert_to_file_path()
+                            req.audio_urls.append(audio_path)
+                            _append_quoted_audio_attachment(req, audio_path)
                         elif isinstance(reply_comp, File):
                             file_path = await reply_comp.get_file()
                             file_name = reply_comp.name or os.path.basename(file_path)
@@ -1127,6 +1372,7 @@ async def build_main_agent(
     if isinstance(req.contexts, str):
         req.contexts = json.loads(req.contexts)
     req.image_urls = normalize_and_dedupe_strings(req.image_urls)
+    req.audio_urls = normalize_and_dedupe_strings(req.audio_urls)
 
     if config.file_extract_enabled:
         try:
@@ -1134,7 +1380,7 @@ async def build_main_agent(
         except Exception as exc:  # noqa: BLE001
             logger.error("Error occurred while applying file extract: %s", exc)
 
-    if not req.prompt and not req.image_urls:
+    if not req.prompt and not req.image_urls and not req.audio_urls:
         if not event.get_group_id() and req.extra_user_content_parts:
             req.prompt = "<attachment>"
         else:
@@ -1149,6 +1395,7 @@ async def build_main_agent(
 
     _modalities_fix(provider, req)
     _plugin_tool_fix(event, req)
+    await _apply_web_search_tools(event, req, plugin_context)
     _sanitize_context_by_modalities(config, provider, req)
 
     if config.llm_safety_mode:
@@ -1157,7 +1404,7 @@ async def build_main_agent(
     if config.computer_use_runtime == "sandbox":
         _apply_sandbox_tools(config, req, req.session_id)
     elif config.computer_use_runtime == "local":
-        _apply_local_env_tools(req)
+        _apply_local_env_tools(req, plugin_context)
 
     agent_runner = AgentRunner()
     astr_agent_ctx = AstrAgentContext(
@@ -1166,12 +1413,16 @@ async def build_main_agent(
     )
 
     if config.add_cron_tools:
-        _proactive_cron_job_tools(req)
+        _proactive_cron_job_tools(req, plugin_context)
 
     if event.platform_meta.support_proactive_message:
         if req.func_tool is None:
             req.func_tool = ToolSet()
-        req.func_tool.add_tool(SEND_MESSAGE_TO_USER_TOOL)
+        req.func_tool.add_tool(
+            plugin_context.get_llm_tool_manager().get_builtin_tool(
+                SendMessageToUserTool
+            )
+        )
 
     if provider.provider_config.get("max_context_tokens", 0) <= 0:
         model = provider.get_model()
@@ -1189,6 +1440,15 @@ async def build_main_agent(
             if config.tool_schema_mode == "full"
             else TOOL_CALL_PROMPT_SKILLS_LIKE_MODE
         )
+
+        if config.computer_use_runtime == "local":
+            tool_prompt += (
+                f"\nCurrent workspace you can use: "
+                f"`{_get_workspace_path_for_umo(event.unified_msg_origin)}`\n"
+                "Unless the user explicitly specifies a different directory, "
+                "perform all file-related operations in this workspace.\n"
+            )
+
         req.system_prompt += f"\n{tool_prompt}\n"
 
     action_type = event.get_extra("action_type")
@@ -1214,6 +1474,14 @@ async def build_main_agent(
         fallback_providers=_get_fallback_chat_providers(
             provider, plugin_context, config.provider_settings
         ),
+        tool_result_overflow_dir=(
+            get_astrbot_system_tmp_path()
+            if req.func_tool and req.func_tool.get_tool("astrbot_file_read_tool")
+            else None
+        ),
+        read_tool=(
+            req.func_tool.get_tool("astrbot_file_read_tool") if req.func_tool else None
+        ),
     )
 
     if apply_reset:

astrbot/core/astr_main_agent_resources.py

@@ -1,42 +1,4 @@
 import base64
-import json
-import os
-import uuid
-
-from pydantic import Field
-from pydantic.dataclasses import dataclass
-
-import astrbot.core.message.components as Comp
-from astrbot.api import logger, sp
-from astrbot.core.agent.run_context import ContextWrapper
-from astrbot.core.agent.tool import FunctionTool, ToolExecResult
-from astrbot.core.astr_agent_context import AstrAgentContext
-from astrbot.core.computer.computer_client import get_booter
-from astrbot.core.computer.tools import (
-    AnnotateExecutionTool,
-    BrowserBatchExecTool,
-    BrowserExecTool,
-    CreateSkillCandidateTool,
-    CreateSkillPayloadTool,
-    EvaluateSkillCandidateTool,
-    ExecuteShellTool,
-    FileDownloadTool,
-    FileUploadTool,
-    GetExecutionHistoryTool,
-    GetSkillPayloadTool,
-    ListSkillCandidatesTool,
-    ListSkillReleasesTool,
-    LocalPythonTool,
-    PromoteSkillCandidateTool,
-    PythonTool,
-    RollbackSkillReleaseTool,
-    RunBrowserSkillTool,
-    SyncSkillReleaseTool,
-)
-from astrbot.core.message.message_event_result import MessageChain
-from astrbot.core.platform.message_session import MessageSession
-from astrbot.core.star.context import Context
-from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 
 LLM_SAFETY_MODE_SYSTEM_PROMPT = """You are running in Safe Mode.
 
@@ -146,356 +108,6 @@ BACKGROUND_TASK_RESULT_WOKE_SYSTEM_PROMPT = (
     "{background_task_result}"
 )
 
-
-@dataclass
-class KnowledgeBaseQueryTool(FunctionTool[AstrAgentContext]):
-    name: str = "astr_kb_search"
-    description: str = (
-        "Query the knowledge base for facts or relevant context. "
-        "Use this tool when the user's question requires factual information, "
-        "definitions, background knowledge, or previously indexed content. "
-        "Only send short keywords or a concise question as the query."
-    )
-    parameters: dict = Field(
-        default_factory=lambda: {
-            "type": "object",
-            "properties": {
-                "query": {
-                    "type": "string",
-                    "description": "A concise keyword query for the knowledge base.",
-                },
-            },
-            "required": ["query"],
-        }
-    )
-
-    async def call(
-        self, context: ContextWrapper[AstrAgentContext], **kwargs
-    ) -> ToolExecResult:
-        query = kwargs.get("query", "")
-        if not query:
-            return "error: Query parameter is empty."
-        result = await retrieve_knowledge_base(
-            query=kwargs.get("query", ""),
-            umo=context.context.event.unified_msg_origin,
-            context=context.context.context,
-        )
-        if not result:
-            return "No relevant knowledge found."
-        return result
-
-
-@dataclass
-class SendMessageToUserTool(FunctionTool[AstrAgentContext]):
-    name: str = "send_message_to_user"
-    description: str = (
-        "Send message to the user. "
-        "Supports various message types including `plain`, `image`, `record`, `video`, `file`, and `mention_user`. "
-        "Use this tool to send media files (`image`, `record`, `video`, `file`), "
-        "or when you need to proactively message the user(such as cron job). For normal text replies, you can output directly."
-    )
-
-    parameters: dict = Field(
-        default_factory=lambda: {
-            "type": "object",
-            "properties": {
-                "messages": {
-                    "type": "array",
-                    "description": "An ordered list of message components to send. `mention_user` type can be used to mention the user.",
-                    "items": {
-                        "type": "object",
-                        "properties": {
-                            "type": {
-                                "type": "string",
-                                "description": (
-                                    "Component type. One of: "
-                                    "plain, image, record, video, file, mention_user. Record is voice message."
-                                ),
-                            },
-                            "text": {
-                                "type": "string",
-                                "description": "Text content for `plain` type.",
-                            },
-                            "path": {
-                                "type": "string",
-                                "description": "File path for `image`, `record`, or `file` types. Both local path and sandbox path are supported.",
-                            },
-                            "url": {
-                                "type": "string",
-                                "description": "URL for `image`, `record`, or `file` types.",
-                            },
-                            "mention_user_id": {
-                                "type": "string",
-                                "description": "User ID to mention for `mention_user` type.",
-                            },
-                        },
-                        "required": ["type"],
-                    },
-                },
-            },
-            "required": ["messages"],
-        }
-    )
-
-    async def _resolve_path_from_sandbox(
-        self, context: ContextWrapper[AstrAgentContext], path: str
-    ) -> tuple[str, bool]:
-        """
-        If the path exists locally, return it directly.
-        Otherwise, check if it exists in the sandbox and download it.
-
-        bool: indicates whether the file was downloaded from sandbox.
-        """
-        if os.path.exists(path):
-            return path, False
-
-        # Try to check if the file exists in the sandbox
-        try:
-            sb = await get_booter(
-                context.context.context,
-                context.context.event.unified_msg_origin,
-            )
-            # Use shell to check if the file exists in sandbox
-            result = await sb.shell.exec(f"test -f {path} && echo '_&exists_'")
-            if "_&exists_" in json.dumps(result):
-                # Download the file from sandbox
-                name = os.path.basename(path)
-                local_path = os.path.join(
-                    get_astrbot_temp_path(), f"sandbox_{uuid.uuid4().hex[:4]}_{name}"
-                )
-                await sb.download_file(path, local_path)
-                logger.info(f"Downloaded file from sandbox: {path} -> {local_path}")
-                return local_path, True
-        except Exception as e:
-            logger.warning(f"Failed to check/download file from sandbox: {e}")
-
-        # Return the original path (will likely fail later, but that's expected)
-        return path, False
-
-    async def call(
-        self, context: ContextWrapper[AstrAgentContext], **kwargs
-    ) -> ToolExecResult:
-        session = kwargs.get("session") or context.context.event.unified_msg_origin
-        messages = kwargs.get("messages")
-
-        if not isinstance(messages, list) or not messages:
-            return "error: messages parameter is empty or invalid."
-
-        components: list[Comp.BaseMessageComponent] = []
-
-        for idx, msg in enumerate(messages):
-            if not isinstance(msg, dict):
-                return f"error: messages[{idx}] should be an object."
-
-            msg_type = str(msg.get("type", "")).lower()
-            if not msg_type:
-                return f"error: messages[{idx}].type is required."
-
-            file_from_sandbox = False
-
-            try:
-                if msg_type == "plain":
-                    text = str(msg.get("text", "")).strip()
-                    if not text:
-                        return f"error: messages[{idx}].text is required for plain component."
-                    components.append(Comp.Plain(text=text))
-                elif msg_type == "image":
-                    path = msg.get("path")
-                    url = msg.get("url")
-                    if path:
-                        (
-                            local_path,
-                            file_from_sandbox,
-                        ) = await self._resolve_path_from_sandbox(context, path)
-                        components.append(Comp.Image.fromFileSystem(path=local_path))
-                    elif url:
-                        components.append(Comp.Image.fromURL(url=url))
-                    else:
-                        return f"error: messages[{idx}] must include path or url for image component."
-                elif msg_type == "record":
-                    path = msg.get("path")
-                    url = msg.get("url")
-                    if path:
-                        (
-                            local_path,
-                            file_from_sandbox,
-                        ) = await self._resolve_path_from_sandbox(context, path)
-                        components.append(Comp.Record.fromFileSystem(path=local_path))
-                    elif url:
-                        components.append(Comp.Record.fromURL(url=url))
-                    else:
-                        return f"error: messages[{idx}] must include path or url for record component."
-                elif msg_type == "video":
-                    path = msg.get("path")
-                    url = msg.get("url")
-                    if path:
-                        (
-                            local_path,
-                            file_from_sandbox,
-                        ) = await self._resolve_path_from_sandbox(context, path)
-                        components.append(Comp.Video.fromFileSystem(path=local_path))
-                    elif url:
-                        components.append(Comp.Video.fromURL(url=url))
-                    else:
-                        return f"error: messages[{idx}] must include path or url for video component."
-                elif msg_type == "file":
-                    path = msg.get("path")
-                    url = msg.get("url")
-                    name = (
-                        msg.get("text")
-                        or (os.path.basename(path) if path else "")
-                        or (os.path.basename(url) if url else "")
-                        or "file"
-                    )
-                    if path:
-                        (
-                            local_path,
-                            file_from_sandbox,
-                        ) = await self._resolve_path_from_sandbox(context, path)
-                        components.append(Comp.File(name=name, file=local_path))
-                    elif url:
-                        components.append(Comp.File(name=name, url=url))
-                    else:
-                        return f"error: messages[{idx}] must include path or url for file component."
-                elif msg_type == "mention_user":
-                    mention_user_id = msg.get("mention_user_id")
-                    if not mention_user_id:
-                        return f"error: messages[{idx}].mention_user_id is required for mention_user component."
-                    components.append(
-                        Comp.At(
-                            qq=mention_user_id,
-                        ),
-                    )
-                else:
-                    return (
-                        f"error: unsupported message type '{msg_type}' at index {idx}."
-                    )
-            except Exception as exc:  # 捕获组件构造异常，避免直接抛出
-                return f"error: failed to build messages[{idx}] component: {exc}"
-
-        try:
-            target_session = (
-                MessageSession.from_str(session)
-                if isinstance(session, str)
-                else session
-            )
-        except Exception as e:
-            return f"error: invalid session: {e}"
-
-        await context.context.context.send_message(
-            target_session,
-            MessageChain(chain=components),
-        )
-
-        # if file_from_sandbox:
-        #     try:
-        #         os.remove(local_path)
-        #     except Exception as e:
-        #         logger.error(f"Error removing temp file {local_path}: {e}")
-
-        return f"Message sent to session {target_session}"
-
-
-async def retrieve_knowledge_base(
-    query: str,
-    umo: str,
-    context: Context,
-) -> str | None:
-    """Inject knowledge base context into the provider request
-
-    Args:
-        umo: Unique message object (session ID)
-        p_ctx: Pipeline context
-    """
-    kb_mgr = context.kb_manager
-    config = context.get_config(umo=umo)
-
-    # 1. 优先读取会话级配置
-    session_config = await sp.session_get(umo, "kb_config", default={})
-
-    if session_config and "kb_ids" in session_config:
-        # 会话级配置
-        kb_ids = session_config.get("kb_ids", [])
-
-        # 如果配置为空列表，明确表示不使用知识库
-        if not kb_ids:
-            logger.info(f"[知识库] 会话 {umo} 已被配置为不使用知识库")
-            return
-
-        top_k = session_config.get("top_k", 5)
-
-        # 将 kb_ids 转换为 kb_names
-        kb_names = []
-        invalid_kb_ids = []
-        for kb_id in kb_ids:
-            kb_helper = await kb_mgr.get_kb(kb_id)
-            if kb_helper:
-                kb_names.append(kb_helper.kb.kb_name)
-            else:
-                logger.warning(f"[知识库] 知识库不存在或未加载: {kb_id}")
-                invalid_kb_ids.append(kb_id)
-
-        if invalid_kb_ids:
-            logger.warning(
-                f"[知识库] 会话 {umo} 配置的以下知识库无效: {invalid_kb_ids}",
-            )
-
-        if not kb_names:
-            return
-
-        logger.debug(f"[知识库] 使用会话级配置，知识库数量: {len(kb_names)}")
-    else:
-        kb_names = config.get("kb_names", [])
-        top_k = config.get("kb_final_top_k", 5)
-        logger.debug(f"[知识库] 使用全局配置，知识库数量: {len(kb_names)}")
-
-    top_k_fusion = config.get("kb_fusion_top_k", 20)
-
-    if not kb_names:
-        return
-
-    logger.debug(f"[知识库] 开始检索知识库，数量: {len(kb_names)}, top_k={top_k}")
-    kb_context = await kb_mgr.retrieve(
-        query=query,
-        kb_names=kb_names,
-        top_k_fusion=top_k_fusion,
-        top_m_final=top_k,
-    )
-
-    if not kb_context:
-        return
-
-    formatted = kb_context.get("context_text", "")
-    if formatted:
-        results = kb_context.get("results", [])
-        logger.debug(f"[知识库] 为会话 {umo} 注入了 {len(results)} 条相关知识块")
-        return formatted
-
-
-KNOWLEDGE_BASE_QUERY_TOOL = KnowledgeBaseQueryTool()
-SEND_MESSAGE_TO_USER_TOOL = SendMessageToUserTool()
-
-EXECUTE_SHELL_TOOL = ExecuteShellTool()
-LOCAL_EXECUTE_SHELL_TOOL = ExecuteShellTool(is_local=True)
-PYTHON_TOOL = PythonTool()
-LOCAL_PYTHON_TOOL = LocalPythonTool()
-FILE_UPLOAD_TOOL = FileUploadTool()
-FILE_DOWNLOAD_TOOL = FileDownloadTool()
-BROWSER_EXEC_TOOL = BrowserExecTool()
-BROWSER_BATCH_EXEC_TOOL = BrowserBatchExecTool()
-RUN_BROWSER_SKILL_TOOL = RunBrowserSkillTool()
-GET_EXECUTION_HISTORY_TOOL = GetExecutionHistoryTool()
-ANNOTATE_EXECUTION_TOOL = AnnotateExecutionTool()
-CREATE_SKILL_PAYLOAD_TOOL = CreateSkillPayloadTool()
-GET_SKILL_PAYLOAD_TOOL = GetSkillPayloadTool()
-CREATE_SKILL_CANDIDATE_TOOL = CreateSkillCandidateTool()
-LIST_SKILL_CANDIDATES_TOOL = ListSkillCandidatesTool()
-EVALUATE_SKILL_CANDIDATE_TOOL = EvaluateSkillCandidateTool()
-PROMOTE_SKILL_CANDIDATE_TOOL = PromoteSkillCandidateTool()
-LIST_SKILL_RELEASES_TOOL = ListSkillReleasesTool()
-ROLLBACK_SKILL_RELEASE_TOOL = RollbackSkillReleaseTool()
-SYNC_SKILL_RELEASE_TOOL = SyncSkillReleaseTool()
-
 # we prevent astrbot from connecting to known malicious hosts
 # these hosts are base64 encoded
 BLOCKED = {"dGZid2h2d3IuY2xvdWQuc2VhbG9zLmlv", "a291cmljaGF0"}

astrbot/core/backup/constants.py

@@ -7,6 +7,7 @@ from sqlmodel import SQLModel
 
 from astrbot.core.db.po import (
     Attachment,
+    ChatUIProject,
     CommandConfig,
     CommandConflict,
     ConversationV2,
@@ -16,6 +17,7 @@ from astrbot.core.db.po import (
     PlatformSession,
     PlatformStat,
     Preference,
+    SessionProjectRelation,
 )
 from astrbot.core.knowledge_base.models import (
     KBDocument,
@@ -44,6 +46,8 @@ MAIN_DB_MODELS: dict[str, type[SQLModel]] = {
     "preferences": Preference,
     "platform_message_history": PlatformMessageHistory,
     "platform_sessions": PlatformSession,
+    "chatui_projects": ChatUIProject,
+    "session_project_relations": SessionProjectRelation,
     "attachments": Attachment,
     "command_configs": CommandConfig,
     "command_conflicts": CommandConflict,

astrbot/core/computer/booters/boxlite.py

@@ -4,7 +4,7 @@ from typing import Any
 
 import aiohttp
 import boxlite
-from shipyard.filesystem import FileSystemComponent as ShipyardFileSystemComponent
+from shipyard import FileSystemComponent as ShipyardFileSystemComponent
 from shipyard.python import PythonComponent as ShipyardPythonComponent
 from shipyard.shell import ShellComponent as ShipyardShellComponent
 
@@ -12,6 +12,7 @@ from astrbot.api import logger
 
 from ..olayer import FileSystemComponent, PythonComponent, ShellComponent
 from .base import ComputerBooter
+from .shipyard import ShipyardFileSystemWrapper
 
 
 class MockShipyardSandboxClient:
@@ -150,11 +151,6 @@ class BoxliteBooter(ComputerBooter):
         self.mocked = MockShipyardSandboxClient(
             sb_url=f"http://127.0.0.1:{random_port}"
         )
-        self._fs = ShipyardFileSystemComponent(
-            client=self.mocked,  # type: ignore
-            ship_id=self.box.id,
-            session_id=session_id,
-        )
         self._python = ShipyardPythonComponent(
             client=self.mocked,  # type: ignore
             ship_id=self.box.id,
@@ -165,6 +161,14 @@ class BoxliteBooter(ComputerBooter):
             ship_id=self.box.id,
             session_id=session_id,
         )
+        self._ship_fs = ShipyardFileSystemComponent(
+            client=self.mocked,  # type: ignore
+            ship_id=self.box.id,
+            session_id=session_id,
+        )
+        self._fs = ShipyardFileSystemWrapper(
+            _shipyard_fs=self._ship_fs, _shipyard_shell=self._shell
+        )
 
         await self.mocked.wait_healthy(self.box.id, session_id)
 

astrbot/core/computer/booters/local.py

@@ -9,15 +9,18 @@ import sys
 from dataclasses import dataclass
 from typing import Any
 
+from python_ripgrep import search
+
 from astrbot.api import logger
-from astrbot.core.utils.astrbot_path import (
-    get_astrbot_data_path,
-    get_astrbot_root,
-    get_astrbot_temp_path,
+from astrbot.core.computer.file_read_utils import (
+    detect_text_encoding,
+    read_local_text_range_sync,
 )
+from astrbot.core.utils.astrbot_path import get_astrbot_root
 
 from ..olayer import FileSystemComponent, PythonComponent, ShellComponent
 from .base import ComputerBooter
+from .shipyard_search_file_util import _truncate_long_lines
 
 _BLOCKED_COMMAND_PATTERNS = [
     " rm -rf ",
@@ -41,43 +44,45 @@ def _is_safe_command(command: str) -> bool:
     return not any(pat in cmd for pat in _BLOCKED_COMMAND_PATTERNS)
 
 
-def _ensure_safe_path(path: str) -> str:
-    abs_path = os.path.abspath(path)
-    allowed_roots = [
-        os.path.abspath(get_astrbot_root()),
-        os.path.abspath(get_astrbot_data_path()),
-        os.path.abspath(get_astrbot_temp_path()),
-    ]
-    if not any(abs_path.startswith(root) for root in allowed_roots):
-        raise PermissionError("Path is outside the allowed computer roots.")
-    return abs_path
-
-
-def _decode_shell_output(output: bytes | None) -> str:
+def _decode_bytes_with_fallback(
+    output: bytes | None,
+    *,
+    preferred_encoding: str | None = None,
+) -> str:
     if output is None:
         return ""
 
     preferred = locale.getpreferredencoding(False) or "utf-8"
-    try:
-        return output.decode("utf-8")
-    except (LookupError, UnicodeDecodeError):
-        pass
+    attempted_encodings: list[str] = []
+
+    def _try_decode(encoding: str) -> str | None:
+        normalized = encoding.lower()
+        if normalized in attempted_encodings:
+            return None
+        attempted_encodings.append(normalized)
+        try:
+            return output.decode(encoding)
+        except (LookupError, UnicodeDecodeError):
+            return None
+
+    for encoding in filter(None, [preferred_encoding, "utf-8", "utf-8-sig"]):
+        if decoded := _try_decode(encoding):
+            return decoded
 
     if os.name == "nt":
-        for encoding in ("mbcs", "cp936", "gbk", "gb18030"):
-            try:
-                return output.decode(encoding)
-            except (LookupError, UnicodeDecodeError):
-                continue
-
-    try:
-        return output.decode(preferred)
-    except (LookupError, UnicodeDecodeError):
-        pass
+        for encoding in ("mbcs", "cp936", "gbk", "gb18030", preferred):
+            if decoded := _try_decode(encoding):
+                return decoded
+    elif decoded := _try_decode(preferred):
+        return decoded
 
     return output.decode("utf-8", errors="replace")
 
 
+def _decode_shell_output(output: bytes | None) -> str:
+    return _decode_bytes_with_fallback(output, preferred_encoding="utf-8")
+
+
 @dataclass
 class LocalShellComponent(ShellComponent):
     async def exec(
@@ -96,7 +101,7 @@ class LocalShellComponent(ShellComponent):
             run_env = os.environ.copy()
             if env:
                 run_env.update({str(k): str(v) for k, v in env.items()})
-            working_dir = _ensure_safe_path(cwd) if cwd else get_astrbot_root()
+            working_dir = os.path.abspath(cwd) if cwd else get_astrbot_root()
             if background:
                 # `command` is intentionally executed through the current shell so
                 # local computer-use behavior matches existing tool semantics.
@@ -172,7 +177,7 @@ class LocalFileSystemComponent(FileSystemComponent):
         self, path: str, content: str = "", mode: int = 0o644
     ) -> dict[str, Any]:
         def _run() -> dict[str, Any]:
-            abs_path = _ensure_safe_path(path)
+            abs_path = os.path.abspath(path)
             os.makedirs(os.path.dirname(abs_path), exist_ok=True)
             with open(abs_path, "w", encoding="utf-8") as f:
                 f.write(content)
@@ -181,12 +186,85 @@ class LocalFileSystemComponent(FileSystemComponent):
 
         return await asyncio.to_thread(_run)
 
-    async def read_file(self, path: str, encoding: str = "utf-8") -> dict[str, Any]:
+    async def read_file(
+        self,
+        path: str,
+        encoding: str = "utf-8",
+        offset: int | None = None,
+        limit: int | None = None,
+    ) -> dict[str, Any]:
         def _run() -> dict[str, Any]:
-            abs_path = _ensure_safe_path(path)
+            abs_path = os.path.abspath(path)
+            detected_encoding = encoding
+            if encoding == "utf-8":
+                with open(abs_path, "rb") as f:
+                    raw_sample = f.read(8192)
+                detected_encoding = detect_text_encoding(raw_sample) or encoding
+            return {
+                "success": True,
+                "content": read_local_text_range_sync(
+                    abs_path,
+                    encoding=detected_encoding,
+                    offset=offset,
+                    limit=limit,
+                ),
+            }
+
+        return await asyncio.to_thread(_run)
+
+    async def search_files(
+        self,
+        pattern: str,
+        path: str | None = None,
+        glob: str | None = None,
+        after_context: int | None = None,
+        before_context: int | None = None,
+    ) -> dict[str, Any]:
+        def _run() -> dict[str, Any]:
+            results = search(
+                patterns=[pattern],
+                paths=[path] if path else None,
+                globs=[glob] if glob else None,
+                after_context=after_context,
+                before_context=before_context,
+                line_number=True,
+            )
+            return {"success": True, "content": _truncate_long_lines("".join(results))}
+
+        return await asyncio.to_thread(_run)
+
+    async def edit_file(
+        self,
+        path: str,
+        old_string: str,
+        new_string: str,
+        replace_all: bool = False,
+        encoding: str = "utf-8",
+    ) -> dict[str, Any]:
+        def _run() -> dict[str, Any]:
+            abs_path = os.path.abspath(path)
             with open(abs_path, encoding=encoding) as f:
                 content = f.read()
-            return {"success": True, "content": content}
+            occurrences = content.count(old_string)
+            if occurrences == 0:
+                return {
+                    "success": False,
+                    "error": "old string not found in file",
+                    "replacements": 0,
+                }
+            if replace_all:
+                updated = content.replace(old_string, new_string)
+                replacements = occurrences
+            else:
+                updated = content.replace(old_string, new_string, 1)
+                replacements = 1
+            with open(abs_path, "w", encoding=encoding) as f:
+                f.write(updated)
+            return {
+                "success": True,
+                "path": abs_path,
+                "replacements": replacements,
+            }
 
         return await asyncio.to_thread(_run)
 
@@ -194,7 +272,7 @@ class LocalFileSystemComponent(FileSystemComponent):
         self, path: str, content: str, mode: str = "w", encoding: str = "utf-8"
     ) -> dict[str, Any]:
         def _run() -> dict[str, Any]:
-            abs_path = _ensure_safe_path(path)
+            abs_path = os.path.abspath(path)
             os.makedirs(os.path.dirname(abs_path), exist_ok=True)
             with open(abs_path, mode, encoding=encoding) as f:
                 f.write(content)
@@ -204,7 +282,7 @@ class LocalFileSystemComponent(FileSystemComponent):
 
     async def delete_file(self, path: str) -> dict[str, Any]:
         def _run() -> dict[str, Any]:
-            abs_path = _ensure_safe_path(path)
+            abs_path = os.path.abspath(path)
             if os.path.isdir(abs_path):
                 shutil.rmtree(abs_path)
             else:
@@ -217,7 +295,7 @@ class LocalFileSystemComponent(FileSystemComponent):
         self, path: str = ".", show_hidden: bool = False
     ) -> dict[str, Any]:
         def _run() -> dict[str, Any]:
-            abs_path = _ensure_safe_path(path)
+            abs_path = os.path.abspath(path)
             entries = os.listdir(abs_path)
             if not show_hidden:
                 entries = [e for e in entries if not e.startswith(".")]

astrbot/core/computer/booters/shipyard.py

@@ -1,9 +1,87 @@
+from __future__ import annotations
+
+from typing import Any
+
+from shipyard import FileSystemComponent as ShipyardFileSystemComponent
 from shipyard import ShipyardClient, Spec
 
 from astrbot.api import logger
 
 from ..olayer import FileSystemComponent, PythonComponent, ShellComponent
 from .base import ComputerBooter
+from .shipyard_search_file_util import search_files_via_shell
+
+
+class ShipyardFileSystemWrapper:
+    def __init__(
+        self, _shipyard_fs: ShipyardFileSystemComponent, _shipyard_shell: ShellComponent
+    ):
+        self._fs = _shipyard_fs
+        self._shell = _shipyard_shell
+
+    async def create_file(
+        self, path: str, content: str = "", mode: int = 420
+    ) -> dict[str, Any]:
+        return await self._fs.create_file(path=path, content=content, mode=mode)
+
+    async def read_file(
+        self,
+        path: str,
+        encoding: str = "utf-8",
+        offset: int | None = None,
+        limit: int | None = None,
+    ) -> dict[str, Any]:
+        return await self._fs.read_file(
+            path=path, encoding=encoding, offset=offset, limit=limit
+        )
+
+    async def write_file(
+        self, path: str, content: str, mode: str = "w", encoding: str = "utf-8"
+    ) -> dict[str, Any]:
+        return await self._fs.write_file(
+            path=path, content=content, mode=mode, encoding=encoding
+        )
+
+    async def list_dir(
+        self, path: str = ".", show_hidden: bool = False
+    ) -> dict[str, Any]:
+        return await self._fs.list_dir(path=path, show_hidden=show_hidden)
+
+    async def delete_file(self, path: str) -> dict[str, Any]:
+        return await self._fs.delete_file(path=path)
+
+    async def search_files(
+        self,
+        pattern: str,
+        path: str | None = None,
+        glob: str | None = None,
+        after_context: int | None = None,
+        before_context: int | None = None,
+    ) -> dict[str, Any]:
+        return await search_files_via_shell(
+            self._shell,
+            pattern=pattern,
+            path=path,
+            glob=glob,
+            after_context=after_context,
+            before_context=before_context,
+        )
+
+    async def edit_file(
+        self,
+        path: str,
+        old_string: str,
+        new_string: str,
+        replace_all: bool = False,
+        encoding: str = "utf-8",
+    ) -> dict[str, Any]:
+        return await self._fs.edit_file(
+            path=path,
+            old_string=old_string,
+            new_string=new_string,
+            replace_all=replace_all,
+            encoding=encoding,
+        )
 
 
 class ShipyardBooter(ComputerBooter):
@@ -29,13 +107,14 @@ class ShipyardBooter(ComputerBooter):
         )
         logger.info(f"Got sandbox ship: {ship.id} for session: {session_id}")
         self._ship = ship
+        self._fs = ShipyardFileSystemWrapper(self._ship.fs, self._ship.shell)
 
     async def shutdown(self) -> None:
         logger.info("[Computer] Shipyard booter shutdown.")
 
     @property
     def fs(self) -> FileSystemComponent:
-        return self._ship.fs
+        return self._fs
 
     @property
     def python(self) -> PythonComponent:

astrbot/core/computer/booters/shipyard_neo.py

@@ -13,6 +13,15 @@ from ..olayer import (
     ShellComponent,
 )
 from .base import ComputerBooter
+from .shipyard_search_file_util import search_files_via_shell
+
+try:
+    from shipyard_neo import BayClient
+    from shipyard_neo.sandbox import Sandbox
+except ImportError:
+    logger.warning(
+        "shipyard_neo_sdk is not installed. ShipyardNeoBooter will not work without it."
+    )
 
 
 def _maybe_model_dump(value: Any) -> dict[str, Any]:
@@ -25,8 +34,20 @@ def _maybe_model_dump(value: Any) -> dict[str, Any]:
     return {}
 
 
+def _slice_content_by_lines(
+    content: str,
+    *,
+    offset: int | None = None,
+    limit: int | None = None,
+) -> str:
+    lines = content.splitlines(keepends=True)
+    start = 0 if offset is None else offset
+    selected = lines[start:] if limit is None else lines[start : start + limit]
+    return "".join(selected)
+
+
 class NeoPythonComponent(PythonComponent):
-    def __init__(self, sandbox: Any) -> None:
+    def __init__(self, sandbox: Sandbox) -> None:
         self._sandbox = sandbox
 
     async def exec(
@@ -67,7 +88,7 @@ class NeoPythonComponent(PythonComponent):
 
 
 class NeoShellComponent(ShellComponent):
-    def __init__(self, sandbox: Any) -> None:
+    def __init__(self, sandbox: Sandbox) -> None:
         self._sandbox = sandbox
 
     async def exec(
@@ -136,8 +157,9 @@ class NeoShellComponent(ShellComponent):
 
 
 class NeoFileSystemComponent(FileSystemComponent):
-    def __init__(self, sandbox: Any) -> None:
+    def __init__(self, sandbox: Sandbox, shell: ShellComponent) -> None:
         self._sandbox = sandbox
+        self._shell = shell
 
     async def create_file(
         self,
@@ -149,10 +171,71 @@ class NeoFileSystemComponent(FileSystemComponent):
         await self._sandbox.filesystem.write_file(path, content)
         return {"success": True, "path": path}
 
-    async def read_file(self, path: str, encoding: str = "utf-8") -> dict[str, Any]:
+    async def read_file(
+        self,
+        path: str,
+        encoding: str = "utf-8",
+        offset: int | None = None,
+        limit: int | None = None,
+    ) -> dict[str, Any]:
         _ = encoding
         content = await self._sandbox.filesystem.read_file(path)
-        return {"success": True, "path": path, "content": content}
+        return {
+            "success": True,
+            "path": path,
+            "content": _slice_content_by_lines(
+                content,
+                offset=offset,
+                limit=limit,
+            ),
+        }
+
+    async def search_files(
+        self,
+        pattern: str,
+        path: str | None = None,
+        glob: str | None = None,
+        after_context: int | None = None,
+        before_context: int | None = None,
+    ) -> dict[str, Any]:
+        return await search_files_via_shell(
+            self._shell,
+            pattern=pattern,
+            path=path,
+            glob=glob,
+            after_context=after_context,
+            before_context=before_context,
+        )
+
+    async def edit_file(
+        self,
+        path: str,
+        old_string: str,
+        new_string: str,
+        replace_all: bool = False,
+        encoding: str = "utf-8",
+    ) -> dict[str, Any]:
+        _ = encoding
+        content = await self._sandbox.filesystem.read_file(path)
+        occurrences = content.count(old_string)
+        if occurrences == 0:
+            return {
+                "success": False,
+                "error": "old string not found in file",
+                "replacements": 0,
+            }
+        if replace_all:
+            updated = content.replace(old_string, new_string)
+            replacements = occurrences
+        else:
+            updated = content.replace(old_string, new_string, 1)
+            replacements = 1
+        await self._sandbox.filesystem.write_file(path, updated)
+        return {
+            "success": True,
+            "path": path,
+            "replacements": replacements,
+        }
 
     async def write_file(
         self,
@@ -186,7 +269,7 @@ class NeoFileSystemComponent(FileSystemComponent):
 
 
 class NeoBrowserComponent(BrowserComponent):
-    def __init__(self, sandbox: Any) -> None:
+    def __init__(self, sandbox: Sandbox) -> None:
         self._sandbox = sandbox
 
     async def exec(
@@ -271,8 +354,8 @@ class ShipyardNeoBooter(ComputerBooter):
         self._access_token = access_token
         self._profile = profile
         self._ttl = ttl
-        self._client: Any = None
-        self._sandbox: Any = None
+        self._client: BayClient | None = None
+        self._sandbox: Sandbox | None = None
         self._bay_manager: Any = None  # BayContainerManager when auto-started
         self._fs: FileSystemComponent | None = None
         self._python: PythonComponent | None = None
@@ -336,8 +419,6 @@ class ShipyardNeoBooter(ComputerBooter):
                 "or ensure Bay's credentials.json is accessible for auto-discovery."
             )
 
-        from shipyard_neo import BayClient
-
         self._client = BayClient(
             endpoint_url=self._endpoint_url,
             access_token=self._access_token,
@@ -352,9 +433,9 @@ class ShipyardNeoBooter(ComputerBooter):
             ttl=self._ttl,
         )
 
-        self._fs = NeoFileSystemComponent(self._sandbox)
-        self._python = NeoPythonComponent(self._sandbox)
         self._shell = NeoShellComponent(self._sandbox)
+        self._fs = NeoFileSystemComponent(self._sandbox, self._shell)
+        self._python = NeoPythonComponent(self._sandbox)
 
         caps = self.capabilities or ()
         self._browser = (

astrbot/core/computer/booters/shipyard_search_file_util.py

@@ -0,0 +1,148 @@
+from __future__ import annotations
+
+import shlex
+from typing import Any
+
+from ..olayer import ShellComponent
+
+_MAX_SEARCH_LINE_COLUMNS = 1000
+
+
+def _truncate_long_lines(text: str) -> str:
+    output_lines: list[str] = []
+    for line in text.splitlines(keepends=True):
+        line_ending = ""
+        line_body = line
+        if line.endswith("\r\n"):
+            line_body = line[:-2]
+            line_ending = "\r\n"
+        elif line.endswith("\n") or line.endswith("\r"):
+            line_body = line[:-1]
+            line_ending = line[-1]
+
+        if len(line_body) > _MAX_SEARCH_LINE_COLUMNS:
+            line_body = line_body[:_MAX_SEARCH_LINE_COLUMNS]
+
+        output_lines.append(f"{line_body}{line_ending}")
+    return "".join(output_lines)
+
+
+def _build_rg_command(
+    *,
+    pattern: str,
+    path: str,
+    glob: str | None,
+    after_context: int | None,
+    before_context: int | None,
+) -> list[str]:
+    command = [
+        "rg",
+        "--color=never",
+        "-n",
+        "--max-columns",
+        str(_MAX_SEARCH_LINE_COLUMNS),
+        "-e",
+        pattern,
+    ]
+    if glob:
+        command.extend(["-g", glob])
+    if after_context is not None:
+        command.extend(["-A", str(after_context)])
+    if before_context is not None:
+        command.extend(["-B", str(before_context)])
+    command.extend(["--", path])
+    return command
+
+
+def _build_grep_command(
+    *,
+    pattern: str,
+    path: str,
+    glob: str | None,
+    after_context: int | None,
+    before_context: int | None,
+) -> list[str]:
+    command = ["grep", "-R", "-H", "-n", "-e", pattern]
+    if glob:
+        command.append(f"--include={glob}")
+    if after_context is not None:
+        command.extend(["-A", str(after_context)])
+    if before_context is not None:
+        command.extend(["-B", str(before_context)])
+    command.extend(["--", path])
+    return command
+
+
+def _quote_command(command: list[str]) -> str:
+    return " ".join(shlex.quote(part) for part in command)
+
+
+def build_search_command(
+    *,
+    pattern: str,
+    path: str,
+    glob: str | None,
+    after_context: int | None,
+    before_context: int | None,
+) -> str:
+    rg_command = _quote_command(
+        _build_rg_command(
+            pattern=pattern,
+            path=path,
+            glob=glob,
+            after_context=after_context,
+            before_context=before_context,
+        )
+    )
+    grep_command = _quote_command(
+        _build_grep_command(
+            pattern=pattern,
+            path=path,
+            glob=glob,
+            after_context=after_context,
+            before_context=before_context,
+        )
+    )
+    return (
+        "if command -v rg >/dev/null 2>&1; then "
+        f"{rg_command}; "
+        "elif command -v grep >/dev/null 2>&1; then "
+        f"{grep_command}; "
+        "else "
+        "echo 'Neither rg nor grep is available in the sandbox.' >&2; "
+        "exit 127; "
+        "fi"
+    )
+
+
+async def search_files_via_shell(
+    shell: ShellComponent,
+    *,
+    pattern: str,
+    path: str | None = None,
+    glob: str | None = None,
+    after_context: int | None = None,
+    before_context: int | None = None,
+    timeout: int = 30,
+) -> dict[str, Any]:
+    command = build_search_command(
+        pattern=pattern,
+        path=path or ".",
+        glob=glob,
+        after_context=after_context,
+        before_context=before_context,
+    )
+    result = await shell.exec(command, timeout=timeout)
+    stdout = _truncate_long_lines(str(result.get("stdout", "") or ""))
+    stderr = str(result.get("stderr", "") or "")
+    exit_code = result.get("exit_code")
+    if exit_code in (0, None):
+        return {"success": True, "content": stdout}
+    if exit_code == 1:
+        return {"success": True, "content": ""}
+    return {
+        "success": False,
+        "content": "",
+        "error": stderr or f"command exited with code {exit_code}",
+        "exit_code": exit_code,
+    }

astrbot/core/computer/computer_client.py

@@ -213,13 +213,24 @@ def parse_description(text: str) -> str:
             break
     if end_idx is None:
         return ""
-    for line in lines[1:end_idx]:
-        if ":" not in line:
-            continue
-        key, value = line.split(":", 1)
-        if key.strip().lower() == "description":
-            return value.strip().strip('"').strip("'")
-    return ""
+
+    frontmatter = "\\n".join(lines[1:end_idx])
+    try:
+        import yaml
+    except ImportError:
+        return ""
+
+    try:
+        payload = yaml.safe_load(frontmatter) or dict()
+    except yaml.YAMLError:
+        return ""
+    if not isinstance(payload, dict):
+        return ""
+
+    description = payload.get("description", "")
+    if not isinstance(description, str):
+        return ""
+    return description.strip()
 
 
 def load_managed_skills() -> list[str]:

astrbot/core/computer/file_read_utils.py

@@ -0,0 +1,707 @@
+from __future__ import annotations
+
+import base64
+import hashlib
+import io
+import json
+import zipfile
+from asyncio import to_thread
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Literal
+
+import mcp
+
+from astrbot.core.agent.context.token_counter import EstimateTokenCounter
+from astrbot.core.agent.message import Message
+from astrbot.core.agent.tool import ToolExecResult
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
+from astrbot.core.utils.media_utils import (
+    IMAGE_COMPRESS_DEFAULT_MAX_SIZE,
+    IMAGE_COMPRESS_DEFAULT_OPTIMIZE,
+    IMAGE_COMPRESS_DEFAULT_QUALITY,
+    _compress_image_sync,
+)
+
+from .booters.base import ComputerBooter
+
+_MAX_FILE_READ_BYTES = 128 * 1024
+_MAX_FILE_READ_TOKENS = 25_000
+_MAX_TEXT_FILE_FULL_READ_BYTES = 256 * 1024
+_FILE_SNIFF_BYTES = 512
+_TOKEN_COUNTER = EstimateTokenCounter()
+_TEXT_ENCODINGS = (
+    "utf-8-sig",
+    "utf-8",
+    "gb18030",
+    "utf-16",
+    "utf-16-le",
+    "utf-16-be",
+    "utf-32",
+    "utf-32-le",
+    "utf-32-be",
+)
+_UTF_BOMS = (
+    b"\xef\xbb\xbf",
+    b"\xff\xfe",
+    b"\xfe\xff",
+    b"\xff\xfe\x00\x00",
+    b"\x00\x00\xfe\xff",
+)
+_ZIP_MAGIC_PREFIXES = (
+    b"PK\x03\x04",
+    b"PK\x05\x06",
+    b"PK\x07\x08",
+)
+_BINARY_MAGIC_PREFIXES = (
+    b"%PDF-",
+    b"\x1f\x8b",
+    b"7z\xbc\xaf\x27\x1c",
+    b"Rar!\x1a\x07",
+    b"\x7fELF",
+    b"MZ",
+)
+
+
+@dataclass(frozen=True)
+class FileProbe:
+    kind: Literal["text", "image", "binary"]
+    encoding: str | None
+    mime_type: str | None
+    size_bytes: int
+
+
+@dataclass(frozen=True)
+class ParsedDocument:
+    kind: Literal["docx", "pdf"]
+    file_bytes: bytes
+    text: str
+
+
+def _build_probe_script(path: str) -> str:
+    return f"""
+import base64
+import json
+from pathlib import Path
+
+path = Path({path!r})
+with path.open("rb") as file_obj:
+    sample = file_obj.read({_FILE_SNIFF_BYTES})
+print(
+    json.dumps(
+        {{
+            "size_bytes": path.stat().st_size,
+            "sample_b64": base64.b64encode(sample).decode("ascii"),
+        }}
+    )
+)
+""".strip()
+
+
+def _build_text_read_script(
+    path: str,
+    *,
+    encoding: str,
+    offset: int | None,
+    limit: int | None,
+) -> str:
+    start_expr = "0" if offset is None else str(offset)
+    limit_expr = "None" if limit is None else str(limit)
+    return f"""
+import json
+from pathlib import Path
+
+path = Path({path!r})
+start = {start_expr}
+limit = {limit_expr}
+end = None if limit is None else start + limit
+lines = []
+with path.open("r", encoding={encoding!r}, newline="") as file_obj:
+    for index, line in enumerate(file_obj):
+        if index < start:
+            continue
+        if end is not None and index >= end:
+            break
+        lines.append(line)
+content = "".join(lines)
+print(json.dumps({{"content": content}}, ensure_ascii=False))
+""".strip()
+
+
+def _build_image_read_script(path: str) -> str:
+    return f"""
+import base64
+import json
+from pathlib import Path
+
+path = Path({path!r})
+data = path.read_bytes()
+print(
+    json.dumps(
+        {{
+            "size_bytes": len(data),
+            "base64": base64.b64encode(data).decode("ascii"),
+        }}
+    )
+)
+""".strip()
+
+
+def _looks_like_text(decoded: str) -> bool:
+    if not decoded:
+        return True
+
+    disallowed = 0
+    printable = 0
+    for char in decoded:
+        if char in "\n\r\t\f\b":
+            printable += 1
+            continue
+        if char.isprintable():
+            printable += 1
+        code = ord(char)
+        if (0 <= code < 32) or (127 <= code < 160):
+            disallowed += 1
+
+    total = max(len(decoded), 1)
+    return disallowed / total <= 0.02 and printable / total >= 0.85
+
+
+def detect_text_encoding(sample: bytes) -> str | None:
+    if not sample:
+        return "utf-8"
+
+    if b"\x00" in sample and not sample.startswith(_UTF_BOMS):
+        odd_bytes = sample[1::2]
+        even_bytes = sample[0::2]
+        odd_zero_ratio = odd_bytes.count(0) / max(len(odd_bytes), 1)
+        even_zero_ratio = even_bytes.count(0) / max(len(even_bytes), 1)
+        if odd_zero_ratio < 0.8 and even_zero_ratio < 0.8:
+            return None
+
+    for encoding in _TEXT_ENCODINGS:
+        try:
+            decoded = sample.decode(encoding)
+        except UnicodeDecodeError as exc:
+            # Probe samples can end in the middle of a multibyte sequence.
+            # When the decode failure only happens at the sample tail, trim a few
+            # bytes and retry so UTF-8 text is not misclassified as binary.
+            if exc.start >= len(sample) - 4:
+                decoded = ""
+                for trim_bytes in range(1, min(4, len(sample)) + 1):
+                    try:
+                        decoded = sample[:-trim_bytes].decode(encoding)
+                        break
+                    except UnicodeDecodeError:
+                        continue
+                if not decoded:
+                    continue
+            else:
+                continue
+        if _looks_like_text(decoded):
+            return encoding
+
+    return None
+
+
+def read_local_text_range_sync(
+    path: str,
+    *,
+    encoding: str,
+    offset: int | None,
+    limit: int | None,
+) -> str:
+    lines: list[str] = []
+    start = 0 if offset is None else offset
+    end = None if limit is None else start + limit
+    with open(path, encoding=encoding, newline="") as file_obj:
+        for index, line in enumerate(file_obj):
+            if index < start:
+                continue
+            if end is not None and index >= end:
+                break
+            lines.append(line)
+    return "".join(lines)
+
+
+async def read_local_text_range(
+    path: str,
+    *,
+    encoding: str,
+    offset: int | None,
+    limit: int | None,
+) -> str:
+    return await to_thread(
+        read_local_text_range_sync,
+        path,
+        encoding=encoding,
+        offset=offset,
+        limit=limit,
+    )
+
+
+async def _exec_python_json(
+    booter: ComputerBooter,
+    script: str,
+    *,
+    action: str,
+) -> dict:
+    result = await booter.python.exec(script)
+    data = result.get("data") if isinstance(result.get("data"), dict) else {}
+    if not isinstance(data, dict):
+        raise RuntimeError(f"{action} failed: invalid result format")
+    output = data.get("output") if isinstance(data.get("output"), dict) else {}
+    if not isinstance(output, dict):
+        raise RuntimeError(f"{action} failed: invalid output format")
+    error_text = str(data.get("error", "") or result.get("error", "") or "").strip()
+    if error_text:
+        raise RuntimeError(f"{action} failed: {error_text}")
+
+    text = str(output.get("text", "") or "").strip()
+    if not text:
+        raise RuntimeError(f"{action} failed: empty output")
+
+    try:
+        payload = json.loads(text)
+    except json.JSONDecodeError as exc:
+        raise RuntimeError(f"{action} failed: invalid JSON output") from exc
+
+    if not isinstance(payload, dict):
+        raise RuntimeError(f"{action} failed: invalid JSON payload")
+    return payload
+
+
+async def _probe_local_file(path: str) -> dict[str, str | int]:
+    def _run() -> dict[str, str | int]:
+        file_path = Path(path)
+        with file_path.open("rb") as file_obj:
+            sample = file_obj.read(_FILE_SNIFF_BYTES)
+        return {
+            "size_bytes": file_path.stat().st_size,
+            "sample_b64": base64.b64encode(sample).decode("ascii"),
+        }
+
+    return await to_thread(_run)
+
+
+async def _read_local_image_base64(path: str) -> dict[str, str | int]:
+    def _run() -> dict[str, str | int]:
+        data = Path(path).read_bytes()
+        return {
+            "size_bytes": len(data),
+            "base64": base64.b64encode(data).decode("ascii"),
+        }
+
+    return await to_thread(_run)
+
+
+async def _read_local_file_bytes(path: str) -> bytes:
+    return await to_thread(Path(path).read_bytes)
+
+
+async def _compress_image_bytes_to_base64(data: bytes) -> dict[str, str | int]:
+    def _run() -> dict[str, str | int]:
+        temp_dir = Path(get_astrbot_temp_path())
+        temp_dir.mkdir(parents=True, exist_ok=True)
+        compressed_path = Path(
+            _compress_image_sync(
+                data,
+                temp_dir,
+                IMAGE_COMPRESS_DEFAULT_MAX_SIZE,
+                IMAGE_COMPRESS_DEFAULT_QUALITY,
+                IMAGE_COMPRESS_DEFAULT_OPTIMIZE,
+            )
+        )
+        try:
+            compressed_bytes = compressed_path.read_bytes()
+        finally:
+            compressed_path.unlink(missing_ok=True)
+
+        return {
+            "size_bytes": len(compressed_bytes),
+            "base64": base64.b64encode(compressed_bytes).decode("ascii"),
+            "mime_type": "image/jpeg",
+        }
+
+    return await to_thread(_run)
+
+
+def _detect_image_mime(sample: bytes) -> str | None:
+    if sample.startswith(b"\x89PNG\r\n\x1a\n"):
+        return "image/png"
+    if sample.startswith(b"\xff\xd8\xff"):
+        return "image/jpeg"
+    if sample.startswith((b"GIF87a", b"GIF89a")):
+        return "image/gif"
+    if sample.startswith(b"BM"):
+        return "image/bmp"
+    if sample.startswith((b"II*\x00", b"MM\x00*")):
+        return "image/tiff"
+    if sample.startswith(b"\x00\x00\x01\x00"):
+        return "image/x-icon"
+    if len(sample) >= 12 and sample[:4] == b"RIFF" and sample[8:12] == b"WEBP":
+        return "image/webp"
+    if len(sample) >= 12 and sample[4:12] in (b"ftypavif", b"ftypavis"):
+        return "image/avif"
+    return None
+
+
+def _looks_like_known_binary(sample: bytes) -> bool:
+    return any(sample.startswith(prefix) for prefix in _BINARY_MAGIC_PREFIXES)
+
+
+def _looks_like_pdf(path: str, sample: bytes) -> bool:
+    return Path(path).suffix.lower() == ".pdf" or sample.startswith(b"%PDF-")
+
+
+def _looks_like_zip_container(sample: bytes) -> bool:
+    return any(sample.startswith(prefix) for prefix in _ZIP_MAGIC_PREFIXES)
+
+
+def _is_docx_bytes(file_bytes: bytes) -> bool:
+    try:
+        with zipfile.ZipFile(io.BytesIO(file_bytes)) as archive:
+            names = set(archive.namelist())
+    except (OSError, zipfile.BadZipFile):
+        return False
+
+    if "[Content_Types].xml" not in names:
+        return False
+
+    return any(name.startswith("word/") for name in names)
+
+
+async def _parse_local_docx_text(file_bytes: bytes, file_name: str) -> str:
+    from astrbot.core.knowledge_base.parsers.markitdown_parser import (
+        MarkitdownParser,
+    )
+
+    result = await MarkitdownParser().parse(file_bytes, file_name)
+    return result.text
+
+
+async def _parse_local_pdf_text(file_bytes: bytes, file_name: str) -> str:
+    from astrbot.core.knowledge_base.parsers.pdf_parser import PDFParser
+
+    result = await PDFParser().parse(file_bytes, file_name)
+    return result.text
+
+
+async def _parse_local_supported_document(
+    path: str,
+    sample: bytes,
+) -> ParsedDocument | None:
+    file_name = Path(path).name
+    if _looks_like_pdf(path, sample):
+        file_bytes = await _read_local_file_bytes(path)
+        text = await _parse_local_pdf_text(file_bytes, file_name)
+        return ParsedDocument(kind="pdf", file_bytes=file_bytes, text=text)
+
+    if Path(path).suffix.lower() == ".docx" or _looks_like_zip_container(sample):
+        file_bytes = await _read_local_file_bytes(path)
+        if not _is_docx_bytes(file_bytes):
+            return None
+        text = await _parse_local_docx_text(file_bytes, file_name)
+        return ParsedDocument(kind="docx", file_bytes=file_bytes, text=text)
+
+    return None
+
+
+def _probe_file(sample: bytes, *, size_bytes: int) -> FileProbe:
+    if image_mime := _detect_image_mime(sample):
+        return FileProbe(
+            kind="image",
+            encoding=None,
+            mime_type=image_mime,
+            size_bytes=size_bytes,
+        )
+
+    if _looks_like_known_binary(sample):
+        return FileProbe(
+            kind="binary",
+            encoding=None,
+            mime_type=None,
+            size_bytes=size_bytes,
+        )
+
+    if encoding := detect_text_encoding(sample):
+        return FileProbe(
+            kind="text",
+            encoding=encoding,
+            mime_type="text/plain",
+            size_bytes=size_bytes,
+        )
+
+    return FileProbe(
+        kind="binary",
+        encoding=None,
+        mime_type=None,
+        size_bytes=size_bytes,
+    )
+
+
+def _validate_text_output(content: str) -> str | None:
+    content_bytes = len(content.encode("utf-8"))
+    if content_bytes > _MAX_FILE_READ_BYTES:
+        return (
+            "Error reading file: "
+            f"output exceeds {_MAX_FILE_READ_BYTES} bytes "
+            f"({content_bytes} bytes). Use `offset`, `limit` to narrow the read window."
+        )
+
+    content_tokens = _TOKEN_COUNTER.count_tokens(
+        [Message(role="user", content=content)]
+    )
+    if content_tokens > _MAX_FILE_READ_TOKENS:
+        return (
+            "Error reading file: "
+            f"output exceeds {_MAX_FILE_READ_TOKENS} tokens "
+            f"({content_tokens} tokens). Use `offset`, `limit` to narrow the read window."
+        )
+
+    return None
+
+
+def _text_exceeds_read_thresholds(content: str) -> bool:
+    return _validate_text_output(content) is not None
+
+
+def _validate_full_text_read_request(probe: FileProbe) -> str | None:
+    if probe.size_bytes > _MAX_TEXT_FILE_FULL_READ_BYTES:
+        return (
+            "Error reading file: "
+            f"text file exceeds {_MAX_TEXT_FILE_FULL_READ_BYTES} bytes "
+            f"({probe.size_bytes} bytes). Use `offset` and `limit` to narrow the read window."
+        )
+    return None
+
+
+def _slice_text_by_lines(
+    content: str,
+    *,
+    offset: int | None,
+    limit: int | None,
+) -> str:
+    if offset is None and limit is None:
+        return content
+
+    lines = content.splitlines(keepends=True)
+    start = 0 if offset is None else offset
+    end = None if limit is None else start + limit
+    return "".join(lines[start:end])
+
+
+async def _store_converted_text_for_workspace(
+    *,
+    workspace_dir: str,
+    original_path: str,
+    original_bytes: bytes,
+    content: str,
+) -> str:
+    def _run() -> str:
+        original_name = Path(original_path).name
+        digest_suffix = hashlib.md5(original_bytes).hexdigest()[-6:]
+        target_dir = (
+            Path(workspace_dir) / "converted_files" / f"{original_name}_{digest_suffix}"
+        )
+        target_dir.mkdir(parents=True, exist_ok=True)
+        target_path = target_dir / "text.txt"
+        target_path.write_text(content, encoding="utf-8")
+        return str(target_path)
+
+    return await to_thread(_run)
+
+
+def _build_converted_text_notice(
+    converted_text_path: str,
+    *,
+    selection_returned: bool,
+    selection_too_large: bool = False,
+) -> str:
+    if selection_too_large:
+        return (
+            "Converted text was saved to "
+            f"`{converted_text_path}`. The requested output is still too large to "
+            "return directly. Read or grep that file with a narrower window."
+        )
+
+    if selection_returned:
+        return (
+            "Full converted text is also available at "
+            f"`{converted_text_path}`. Read or grep that file with a narrow "
+            "window for additional reads."
+        )
+
+    return (
+        "Converted text was saved to "
+        f"`{converted_text_path}` because the parsed document is too large to "
+        "return directly. Read or grep that file with a narrow window."
+    )
+
+
+async def _read_local_supported_document_result(
+    *,
+    path: str,
+    parsed_document: ParsedDocument,
+    workspace_dir: str | None,
+    offset: int | None,
+    limit: int | None,
+) -> ToolExecResult:
+    content = parsed_document.text
+    if not content:
+        return "No content found at the requested line offset."
+
+    if not _text_exceeds_read_thresholds(content):
+        selected_content = _slice_text_by_lines(content, offset=offset, limit=limit)
+        if not selected_content:
+            return "No content found at the requested line offset."
+        if validation_error := _validate_text_output(selected_content):
+            return validation_error
+        return selected_content
+
+    if not workspace_dir:
+        return (
+            "Error reading file: parsed document exceeds the read output limit and "
+            "no workspace is available for storing converted text."
+        )
+
+    converted_text_path = await _store_converted_text_for_workspace(
+        workspace_dir=workspace_dir,
+        original_path=path,
+        original_bytes=parsed_document.file_bytes,
+        content=content,
+    )
+
+    if offset is None and limit is None:
+        return _build_converted_text_notice(
+            converted_text_path,
+            selection_returned=False,
+        )
+
+    selected_content = _slice_text_by_lines(content, offset=offset, limit=limit)
+    if not selected_content:
+        return (
+            "No content found at the requested line offset. "
+            + _build_converted_text_notice(
+                converted_text_path,
+                selection_returned=False,
+            )
+        )
+
+    notice = _build_converted_text_notice(
+        converted_text_path,
+        selection_returned=True,
+    )
+    combined_output = f"{selected_content}\n\n[{notice}]"
+    if _validate_text_output(combined_output):
+        if _validate_text_output(selected_content):
+            return _build_converted_text_notice(
+                converted_text_path,
+                selection_returned=False,
+                selection_too_large=True,
+            )
+        return selected_content
+
+    return combined_output
+
+
+async def read_file_tool_result(
+    booter: ComputerBooter,
+    *,
+    local_mode: bool,
+    path: str,
+    offset: int | None,
+    limit: int | None,
+    workspace_dir: str | None = None,
+) -> ToolExecResult:
+    if local_mode:
+        probe_payload = await _probe_local_file(path)
+    else:
+        probe_payload = await _exec_python_json(
+            booter,
+            _build_probe_script(path),
+            action="file probe",
+        )
+    sample_b64 = str(probe_payload.get("sample_b64", "") or "")
+    sample = base64.b64decode(sample_b64) if sample_b64 else b""
+    size_bytes = int(probe_payload.get("size_bytes", 0) or 0)
+    probe = _probe_file(sample, size_bytes=size_bytes)
+
+    if local_mode:
+        try:
+            parsed_document = await _parse_local_supported_document(path, sample)
+        except Exception as exc:
+            return f"Error reading file: failed to parse document: {exc}"
+
+        if parsed_document is not None:
+            return await _read_local_supported_document_result(
+                path=path,
+                parsed_document=parsed_document,
+                workspace_dir=workspace_dir,
+                offset=offset,
+                limit=limit,
+            )
+
+    if probe.kind == "binary":
+        return "Error reading file: binary files are not supported by this tool."
+
+    if probe.kind == "image":
+        if local_mode:
+            image_payload = await _read_local_image_base64(path)
+        else:
+            image_payload = await _exec_python_json(
+                booter,
+                _build_image_read_script(path),
+                action="image read",
+            )
+        raw_base64_data = str(image_payload.get("base64", "") or "")
+        if not raw_base64_data:
+            return "Error reading file: image payload is empty."
+        raw_bytes = base64.b64decode(raw_base64_data)
+        compressed_payload = await _compress_image_bytes_to_base64(raw_bytes)
+        base64_data = str(compressed_payload.get("base64", "") or "")
+        if not base64_data:
+            return "Error reading file: compressed image payload is empty."
+        return mcp.types.CallToolResult(
+            content=[
+                mcp.types.ImageContent(
+                    type="image",
+                    data=base64_data,
+                    mimeType=str(
+                        compressed_payload.get("mime_type", "") or "image/jpeg"
+                    ),
+                )
+            ]
+        )
+
+    if offset is None and limit is None:
+        if validation_error := _validate_full_text_read_request(probe):
+            return validation_error
+
+    if local_mode:
+        content = await read_local_text_range(
+            path,
+            encoding=probe.encoding or "utf-8",
+            offset=offset,
+            limit=limit,
+        )
+    else:
+        text_payload = await _exec_python_json(
+            booter,
+            _build_text_read_script(
+                path,
+                encoding=probe.encoding or "utf-8",
+                offset=offset,
+                limit=limit,
+            ),
+            action="text read",
+        )
+        content = str(text_payload.get("content", "") or "")
+
+    if not content:
+        return "No content found at the requested line offset."
+
+    if validation_error := _validate_text_output(content):
+        return validation_error
+
+    return content

astrbot/core/computer/olayer/filesystem.py

@@ -12,8 +12,36 @@ class FileSystemComponent(Protocol):
         """Create a file with the specified content"""
         ...
 
-    async def read_file(self, path: str, encoding: str = "utf-8") -> dict[str, Any]:
-        """Read file content"""
+    async def read_file(
+        self,
+        path: str,
+        encoding: str = "utf-8",
+        offset: int | None = None,
+        limit: int | None = None,
+    ) -> dict[str, Any]:
+        """Read file content by line window"""
+        ...
+
+    async def search_files(
+        self,
+        pattern: str,
+        path: str | None = None,
+        glob: str | None = None,
+        after_context: int | None = None,
+        before_context: int | None = None,
+    ) -> dict[str, Any]:
+        """Search file contents"""
+        ...
+
+    async def edit_file(
+        self,
+        path: str,
+        old_string: str,
+        new_string: str,
+        replace_all: bool = False,
+        encoding: str = "utf-8",
+    ) -> dict[str, Any]:
+        """Edit file content by string replacement"""
         ...
 
     async def write_file(

astrbot/core/computer/tools/fs.py

@@ -1,204 +0,0 @@
-import os
-import uuid
-from dataclasses import dataclass, field
-
-from astrbot.api import FunctionTool, logger
-from astrbot.api.event import MessageChain
-from astrbot.core.agent.run_context import ContextWrapper
-from astrbot.core.agent.tool import ToolExecResult
-from astrbot.core.astr_agent_context import AstrAgentContext
-from astrbot.core.message.components import File
-from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
-
-from ..computer_client import get_booter
-from .permissions import check_admin_permission
-
-# @dataclass
-# class CreateFileTool(FunctionTool):
-#     name: str = "astrbot_create_file"
-#     description: str = "Create a new file in the sandbox."
-#     parameters: dict = field(
-#         default_factory=lambda: {
-#             "type": "object",
-#             "properties": {
-#                 "path": {
-#                     "path": "string",
-#                     "description": "The path where the file should be created, relative to the sandbox root. Must not use absolute paths or traverse outside the sandbox.",
-#                 },
-#                 "content": {
-#                     "type": "string",
-#                     "description": "The content to write into the file.",
-#                 },
-#             },
-#             "required": ["path", "content"],
-#         }
-#     )
-
-#     async def call(
-#         self, context: ContextWrapper[AstrAgentContext], path: str, content: str
-#     ) -> ToolExecResult:
-#         sb = await get_booter(
-#             context.context.context,
-#             context.context.event.unified_msg_origin,
-#         )
-#         try:
-#             result = await sb.fs.create_file(path, content)
-#             return json.dumps(result)
-#         except Exception as e:
-#             return f"Error creating file: {str(e)}"
-
-
-# @dataclass
-# class ReadFileTool(FunctionTool):
-#     name: str = "astrbot_read_file"
-#     description: str = "Read the content of a file in the sandbox."
-#     parameters: dict = field(
-#         default_factory=lambda: {
-#             "type": "object",
-#             "properties": {
-#                 "path": {
-#                     "type": "string",
-#                     "description": "The path of the file to read, relative to the sandbox root. Must not use absolute paths or traverse outside the sandbox.",
-#                 },
-#             },
-#             "required": ["path"],
-#         }
-#     )
-
-#     async def call(self, context: ContextWrapper[AstrAgentContext], path: str):
-#         sb = await get_booter(
-#             context.context.context,
-#             context.context.event.unified_msg_origin,
-#         )
-#         try:
-#             result = await sb.fs.read_file(path)
-#             return result
-#         except Exception as e:
-#             return f"Error reading file: {str(e)}"
-
-
-@dataclass
-class FileUploadTool(FunctionTool):
-    name: str = "astrbot_upload_file"
-    description: str = "Upload a local file to the sandbox. The file must exist on the local filesystem."
-    parameters: dict = field(
-        default_factory=lambda: {
-            "type": "object",
-            "properties": {
-                "local_path": {
-                    "type": "string",
-                    "description": "The local file path to upload. This must be an absolute path to an existing file on the local filesystem.",
-                },
-                # "remote_path": {
-                #     "type": "string",
-                #     "description": "The filename to use in the sandbox. If not provided, file will be saved to the working directory with the same name as the local file.",
-                # },
-            },
-            "required": ["local_path"],
-        }
-    )
-
-    async def call(
-        self,
-        context: ContextWrapper[AstrAgentContext],
-        local_path: str,
-    ) -> str | None:
-        if permission_error := check_admin_permission(context, "File upload/download"):
-            return permission_error
-        sb = await get_booter(
-            context.context.context,
-            context.context.event.unified_msg_origin,
-        )
-        try:
-            # Check if file exists
-            if not os.path.exists(local_path):
-                return f"Error: File does not exist: {local_path}"
-
-            if not os.path.isfile(local_path):
-                return f"Error: Path is not a file: {local_path}"
-
-            # Use basename if sandbox_filename is not provided
-            remote_path = os.path.basename(local_path)
-
-            # Upload file to sandbox
-            result = await sb.upload_file(local_path, remote_path)
-            logger.debug(f"Upload result: {result}")
-            success = result.get("success", False)
-
-            if not success:
-                return f"Error uploading file: {result.get('message', 'Unknown error')}"
-
-            file_path = result.get("file_path", "")
-            logger.info(f"File {local_path} uploaded to sandbox at {file_path}")
-
-            return f"File uploaded successfully to {file_path}"
-        except Exception as e:
-            logger.error(f"Error uploading file {local_path}: {e}")
-            return f"Error uploading file: {str(e)}"
-
-
-@dataclass
-class FileDownloadTool(FunctionTool):
-    name: str = "astrbot_download_file"
-    description: str = "Download a file from the sandbox. Only call this when user explicitly need you to download a file."
-    parameters: dict = field(
-        default_factory=lambda: {
-            "type": "object",
-            "properties": {
-                "remote_path": {
-                    "type": "string",
-                    "description": "The path of the file in the sandbox to download.",
-                },
-                "also_send_to_user": {
-                    "type": "boolean",
-                    "description": "Whether to also send the downloaded file to the user via message. Defaults to true.",
-                },
-            },
-            "required": ["remote_path"],
-        }
-    )
-
-    async def call(
-        self,
-        context: ContextWrapper[AstrAgentContext],
-        remote_path: str,
-        also_send_to_user: bool = True,
-    ) -> ToolExecResult:
-        if permission_error := check_admin_permission(context, "File upload/download"):
-            return permission_error
-        sb = await get_booter(
-            context.context.context,
-            context.context.event.unified_msg_origin,
-        )
-        try:
-            name = os.path.basename(remote_path)
-
-            local_path = os.path.join(
-                get_astrbot_temp_path(), f"sandbox_{uuid.uuid4().hex[:4]}_{name}"
-            )
-
-            # Download file from sandbox
-            await sb.download_file(remote_path, local_path)
-            logger.info(f"File {remote_path} downloaded from sandbox to {local_path}")
-
-            if also_send_to_user:
-                try:
-                    name = os.path.basename(local_path)
-                    await context.context.event.send(
-                        MessageChain(chain=[File(name=name, file=local_path)])
-                    )
-                except Exception as e:
-                    logger.error(f"Error sending file message: {e}")
-
-                # remove
-                # try:
-                #     os.remove(local_path)
-                # except Exception as e:
-                #     logger.error(f"Error removing temp file {local_path}: {e}")
-
-                return f"File downloaded successfully to {local_path} and sent to user."
-
-            return f"File downloaded successfully to {local_path}"
-        except Exception as e:
-            logger.error(f"Error downloading file {remote_path}: {e}")
-            return f"Error downloading file: {str(e)}"

astrbot/core/config/astrbot_config.py

@@ -178,4 +178,6 @@ class AstrBotConfig(dict):
         self[key] = value
 
     def check_exist(self) -> bool:
+        if not self.config_path:  # 加判空
+            return False
         return os.path.exists(self.config_path)

astrbot/core/config/default.py

@@ -3,10 +3,43 @@
 import os
 from typing import Any, TypedDict
 
+from astrbot.core.i18n import Language
 from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
-VERSION = "4.20.0"
+VERSION = "4.23.0"
 DB_PATH = os.path.join(get_astrbot_data_path(), "data_v4.db")
+PERSONAL_WECHAT_CONFIG_METADATA = {
+    "weixin_oc_base_url": {
+        "description": "Base URL",
+        "type": "string",
+        "hint": "默认值: https://ilinkai.weixin.qq.com",
+    },
+    "weixin_oc_bot_type": {
+        "description": "扫码参数 bot_type",
+        "type": "string",
+        "hint": "默认值: 3",
+    },
+    "weixin_oc_qr_poll_interval": {
+        "description": "二维码状态轮询间隔（秒）",
+        "type": "int",
+        "hint": "每隔多少秒轮询一次二维码状态。",
+    },
+    "weixin_oc_long_poll_timeout_ms": {
+        "description": "getUpdates 长轮询超时时间（毫秒）",
+        "type": "int",
+        "hint": "会话消息拉取接口超时参数。",
+    },
+    "weixin_oc_api_timeout_ms": {
+        "description": "HTTP 请求超时（毫秒）",
+        "type": "int",
+        "hint": "通用 API 请求超时参数。",
+    },
+    "weixin_oc_token": {
+        "description": "登录后 token（可留空）",
+        "type": "string",
+        "hint": "扫码登录成功后会自动写入；高级场景可手动填写。",
+    },
+}
 
 WEBHOOK_SUPPORTED_PLATFORMS = [
     "qq_official_webhook",
@@ -21,6 +54,7 @@ WEBHOOK_SUPPORTED_PLATFORMS = [
 # 默认配置
 DEFAULT_CONFIG = {
     "config_version": 2,
+    "language": Language.ZH_CN.value,
     "platform_settings": {
         "unique_session": False,
         "rate_limit": {
@@ -74,9 +108,10 @@ DEFAULT_CONFIG = {
         "provider_pool": ["*"],  # "*" 表示使用所有可用的提供者
         "wake_prefix": "",
         "web_search": False,
-        "websearch_provider": "default",
+        "websearch_provider": "tavily",
         "websearch_tavily_key": [],
         "websearch_bocha_key": [],
+        "websearch_brave_key": [],
         "websearch_baidu_app_builder_key": "",
         "web_search_link": False,
         "display_reasoning_text": False,
@@ -117,7 +152,7 @@ DEFAULT_CONFIG = {
         "unsupported_streaming_strategy": "realtime_segmenting",
         "reachability_check": False,
         "max_agent_step": 30,
-        "tool_call_timeout": 60,
+        "tool_call_timeout": 120,
         "tool_schema_mode": "full",
         "llm_safety_mode": True,
         "safety_mode_strategy": "system_prompt",  # TODO: llm judge
@@ -142,6 +177,11 @@ DEFAULT_CONFIG = {
             "shipyard_neo_profile": "python-default",
             "shipyard_neo_ttl": 3600,
         },
+        "image_compress_enabled": True,
+        "image_compress_options": {
+            "max_size": 1280,
+            "quality": 95,
+        },
     },
     # SubAgent orchestrator mode:
     # - main_enable = False: disabled; main LLM mounts tools normally (persona selection).
@@ -364,6 +404,16 @@ CONFIG_METADATA_2 = {
                         "callback_server_host": "0.0.0.0",
                         "port": 6198,
                     },
+                    "个人微信": {
+                        "id": "weixin_personal",
+                        "type": "weixin_oc",
+                        "enable": False,
+                        "weixin_oc_base_url": "https://ilinkai.weixin.qq.com",
+                        "weixin_oc_bot_type": "3",
+                        "weixin_oc_qr_poll_interval": 1,
+                        "weixin_oc_long_poll_timeout_ms": 35_000,
+                        "weixin_oc_api_timeout_ms": 15_000,
+                    },
                     "飞书(Lark)": {
                         "id": "lark",
                         "type": "lark",
@@ -396,6 +446,7 @@ CONFIG_METADATA_2 = {
                         "telegram_command_register": True,
                         "telegram_command_auto_refresh": True,
                         "telegram_command_register_interval": 300,
+                        "telegram_polling_restart_delay": 5.0,
                     },
                     "Discord": {
                         "id": "discord",
@@ -405,6 +456,7 @@ CONFIG_METADATA_2 = {
                         "discord_proxy": "",
                         "discord_command_register": True,
                         "discord_activity_name": "",
+                        "discord_allow_bot_messages": False,
                     },
                     "Misskey": {
                         "id": "misskey",
@@ -458,12 +510,11 @@ CONFIG_METADATA_2 = {
                         "satori_heartbeat_interval": 10,
                         "satori_reconnect_delay": 5,
                     },
-                    "kook": {
+                    "KOOK": {
                         "id": "kook",
                         "type": "kook",
                         "enable": False,
                         "kook_bot_token": "",
-                        "kook_bot_nickname": "",
                         "kook_reconnect_delay": 1,
                         "kook_max_reconnect_delay": 60,
                         "kook_max_retry_delay": 60,
@@ -472,6 +523,14 @@ CONFIG_METADATA_2 = {
                         "kook_max_heartbeat_failures": 3,
                         "kook_max_consecutive_failures": 5,
                     },
+                    "Mattermost": {
+                        "id": "mattermost",
+                        "type": "mattermost",
+                        "enable": False,
+                        "mattermost_url": "https://chat.example.com",
+                        "mattermost_bot_token": "",
+                        "mattermost_reconnect_delay": 5.0,
+                    },
                     # "WebChat": {
                     #     "id": "webchat",
                     #     "type": "webchat",
@@ -606,6 +665,21 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "hint": "如果你的网络环境为中国大陆，请在 `其他配置` 处设置代理或更改 api_base。",
                     },
+                    "mattermost_url": {
+                        "description": "Mattermost URL",
+                        "type": "string",
+                        "hint": "Mattermost 服务地址，例如 https://chat.example.com。",
+                    },
+                    "mattermost_bot_token": {
+                        "description": "Mattermost Bot Token",
+                        "type": "string",
+                        "hint": "在 Mattermost 中创建 Bot 账户后生成的访问令牌。",
+                    },
+                    "mattermost_reconnect_delay": {
+                        "description": "Mattermost 重连延迟",
+                        "type": "float",
+                        "hint": "WebSocket 断开后的重连等待时间，单位为秒。默认 5 秒。",
+                    },
                     "misskey_instance_url": {
                         "description": "Misskey 实例 URL",
                         "type": "string",
@@ -687,6 +761,11 @@ CONFIG_METADATA_2 = {
                         "type": "int",
                         "hint": "Telegram 命令自动刷新间隔，单位为秒。",
                     },
+                    "telegram_polling_restart_delay": {
+                        "description": "Telegram 轮询重启延迟",
+                        "type": "float",
+                        "hint": "当轮询意外结束尝试自动重启时的延迟时间，理论上越短恢复越快，但过短（<0.1s）可能导致死循环针对 API 服务器的请求阻断。单位为秒。默认为 5s。",
+                    },
                     "id": {
                         "description": "机器人名称",
                         "type": "string",
@@ -843,6 +922,11 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "hint": "可选的 Discord 活动名称。留空则不设置活动。",
                     },
+                    "discord_allow_bot_messages": {
+                        "description": "允许接收机器人消息",
+                        "type": "bool",
+                        "hint": "启用后，AstrBot 将接收来自其他 Discord 机器人的消息。适用于机器人间通信场景（如消息转发）。默认关闭。",
+                    },
                     "port": {
                         "description": "回调服务器端口",
                         "type": "int",
@@ -864,6 +948,7 @@ CONFIG_METADATA_2 = {
                         "type": "bool",
                         "hint": "Webhook 模式下使用 AstrBot 统一 Webhook 入口，无需单独开启端口。回调地址为 /api/platform/webhook/{webhook_uuid}。",
                     },
+                    **PERSONAL_WECHAT_CONFIG_METADATA,
                     "webhook_uuid": {
                         "invisible": True,
                         "description": "Webhook UUID",
@@ -875,11 +960,6 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "hint": "必填项。从 KOOK 开发者平台获取的机器人 Token。",
                     },
-                    "kook_bot_nickname": {
-                        "description": "Bot Nickname",
-                        "type": "string",
-                        "hint": "可选项。若发送者昵称与此值一致，将忽略该消息以避免广播风暴。",
-                    },
                     "kook_reconnect_delay": {
                         "description": "重连延迟",
                         "type": "int",
@@ -1074,7 +1154,7 @@ CONFIG_METADATA_2 = {
                 "type": "list",
                 # provider sources templates
                 "config_template": {
-                    "OpenAI": {
+                    "OpenAI Compatible": {
                         "id": "openai",
                         "provider": "openai",
                         "type": "openai_chat_completion",
@@ -1118,6 +1198,20 @@ CONFIG_METADATA_2 = {
                         "api_base": "https://api.anthropic.com/v1",
                         "timeout": 120,
                         "proxy": "",
+                        "custom_headers": {},
+                        "anth_thinking_config": {"type": "", "budget": 0, "effort": ""},
+                    },
+                    "Kimi Coding Plan": {
+                        "id": "kimi-code",
+                        "provider": "kimi-code",
+                        "type": "kimi_code_chat_completion",
+                        "provider_type": "chat_completion",
+                        "enable": True,
+                        "key": [],
+                        "api_base": "https://api.kimi.com/coding/",
+                        "timeout": 120,
+                        "proxy": "",
+                        "custom_headers": {"User-Agent": "claude-code/0.1.0"},
                         "anth_thinking_config": {"type": "", "budget": 0, "effort": ""},
                     },
                     "Moonshot": {
@@ -1181,6 +1275,18 @@ CONFIG_METADATA_2 = {
                         "proxy": "",
                         "custom_headers": {},
                     },
+                    "LongCat": {
+                        "id": "longcat",
+                        "provider": "longcat",
+                        "type": "longcat_chat_completion",
+                        "provider_type": "chat_completion",
+                        "enable": True,
+                        "key": [],
+                        "api_base": "https://api.longcat.chat/openai",
+                        "timeout": 120,
+                        "proxy": "",
+                        "custom_headers": {},
+                    },
                     "AIHubMix": {
                         "id": "aihubmix",
                         "provider": "aihubmix",
@@ -1240,6 +1346,7 @@ CONFIG_METADATA_2 = {
                         "api_base": "http://127.0.0.1:11434/v1",
                         "proxy": "",
                         "custom_headers": {},
+                        "ollama_disable_thinking": False,
                     },
                     "LM Studio": {
                         "id": "lm_studio",
@@ -1437,6 +1544,20 @@ CONFIG_METADATA_2 = {
                         "model": "whisper-1",
                         "proxy": "",
                     },
+                    "MiMo STT(API)": {
+                        "id": "mimo_stt",
+                        "provider": "mimo",
+                        "type": "mimo_stt_api",
+                        "provider_type": "speech_to_text",
+                        "enable": False,
+                        "api_key": "",
+                        "api_base": "https://api.xiaomimimo.com/v1",
+                        "model": "mimo-v2-omni",
+                        "mimo-stt-system-prompt": "You are a speech transcription assistant. Transcribe the spoken content from the audio exactly and return only the transcription text.",
+                        "mimo-stt-user-prompt": "Please transcribe the content of the audio and return only the transcription text.",
+                        "timeout": "20",
+                        "proxy": "",
+                    },
                     "Whisper(Local)": {
                         "provider": "openai",
                         "type": "openai_whisper_selfhost",
@@ -1444,6 +1565,7 @@ CONFIG_METADATA_2 = {
                         "enable": False,
                         "id": "whisper_selfhost",
                         "model": "tiny",
+                        "whisper_device": "cpu",
                     },
                     "SenseVoice(Local)": {
                         "type": "sensevoice_stt_selfhost",
@@ -1467,6 +1589,23 @@ CONFIG_METADATA_2 = {
                         "timeout": "20",
                         "proxy": "",
                     },
+                    "MiMo TTS(API)": {
+                        "id": "mimo_tts",
+                        "type": "mimo_tts_api",
+                        "provider": "mimo",
+                        "provider_type": "text_to_speech",
+                        "enable": False,
+                        "api_key": "",
+                        "api_base": "https://api.xiaomimimo.com/v1",
+                        "model": "mimo-v2-tts",
+                        "mimo-tts-voice": "mimo_default",
+                        "mimo-tts-format": "wav",
+                        "mimo-tts-style-prompt": "",
+                        "mimo-tts-dialect": "",
+                        "mimo-tts-seed-text": "Hello, MiMo, have you had lunch?",
+                        "timeout": "20",
+                        "proxy": "",
+                    },
                     "Genie TTS": {
                         "id": "genie_tts",
                         "provider": "genie_tts",
@@ -1529,10 +1668,14 @@ CONFIG_METADATA_2 = {
                         "type": "gsvi_tts_api",
                         "provider": "gpt_sovits_inference",
                         "provider_type": "text_to_speech",
-                        "api_base": "http://127.0.0.1:5000",
-                        "character": "",
-                        "emotion": "default",
                         "enable": False,
+                        "api_key": "",
+                        "api_base": "http://127.0.0.1:8000",
+                        "version": "v4",
+                        "character": "",
+                        "prompt_text_lang": "中文",
+                        "emotion": "默认",
+                        "text_lang": "中文",
                         "timeout": 20,
                     },
                     "FishAudio TTS(API)": {
@@ -1663,6 +1806,7 @@ CONFIG_METADATA_2 = {
                         "enable": True,
                         "rerank_api_key": "",
                         "rerank_api_base": "http://127.0.0.1:8000",
+                        "rerank_api_suffix": "/v1/rerank",
                         "rerank_model": "BAAI/bge-reranker-base",
                         "timeout": 20,
                     },
@@ -1691,6 +1835,19 @@ CONFIG_METADATA_2 = {
                         "return_documents": False,
                         "instruct": "",
                     },
+                    "NVIDIA Rerank": {
+                        "id": "nvidia_rerank",
+                        "type": "nvidia_rerank",
+                        "provider": "nvidia",
+                        "provider_type": "rerank",
+                        "enable": True,
+                        "nvidia_rerank_api_key": "",
+                        "nvidia_rerank_api_base": "https://ai.api.nvidia.com/v1/retrieval",
+                        "nvidia_rerank_model": "nv-rerank-qa-mistral-4b:1",
+                        "nvidia_rerank_model_endpoint": "/reranking",
+                        "timeout": 20,
+                        "nvidia_rerank_truncate": "",
+                    },
                     "Xinference STT": {
                         "id": "xinference_stt",
                         "type": "xinference_stt",
@@ -1728,7 +1885,12 @@ CONFIG_METADATA_2 = {
                     "rerank_api_base": {
                         "description": "重排序模型 API Base URL",
                         "type": "string",
-                        "hint": "AstrBot 会在请求时在末尾加上 /v1/rerank。",
+                        "hint": "最终请求路径由 Base URL 和路径后缀拼接而成（默认为 /v1/rerank）。",
+                    },
+                    "rerank_api_suffix": {
+                        "description": "API URL 路径后缀",
+                        "type": "string",
+                        "hint": "追加到 base_url 后的路径，如 /v1/rerank。留空则不追加。",
                     },
                     "rerank_api_key": {
                         "description": "API Key",
@@ -1754,12 +1916,40 @@ CONFIG_METADATA_2 = {
                         "type": "bool",
                         "hint": "如果模型当前未在 Xinference 服务中运行，是否尝试自动启动它。在生产环境中建议关闭。",
                     },
+                    "nvidia_rerank_api_base": {
+                        "description": "API Base URL",
+                        "type": "string",
+                    },
+                    "nvidia_rerank_api_key": {
+                        "description": "API Key",
+                        "type": "string",
+                    },
+                    "nvidia_rerank_model": {
+                        "description": "重排序模型名称",
+                        "type": "string",
+                        "hint": "请参照NVIDIA Docs中模型名称填写。",
+                    },
+                    "nvidia_rerank_model_endpoint": {
+                        "description": "自定义模型端点",
+                        "type": "string",
+                        "hint": "自定义URL末尾端点，默认为 /reranking",
+                    },
+                    "nvidia_rerank_truncate": {
+                        "description": "文本截断策略",
+                        "type": "string",
+                        "hint": "当输入文本过长时，是否截断输入以适应模型的最大上下文长度。",
+                        "options": [
+                            "",
+                            "NONE",
+                            "END",
+                        ],
+                    },
                     "modalities": {
                         "description": "模型能力",
                         "type": "list",
                         "items": {"type": "string"},
-                        "options": ["text", "image", "tool_use"],
-                        "labels": ["文本", "图像", "工具使用"],
+                        "options": ["text", "image", "audio", "tool_use"],
+                        "labels": ["文本", "图像", "音频", "工具使用"],
                         "render_type": "checkbox",
                         "hint": "模型支持的模态。如所填写的模型不支持图像，请取消勾选图像。",
                     },
@@ -1769,6 +1959,11 @@ CONFIG_METADATA_2 = {
                         "items": {},
                         "hint": "此处添加的键值对将被合并到 OpenAI SDK 的 default_headers 中，用于自定义 HTTP 请求头。值必须为字符串。",
                     },
+                    "ollama_disable_thinking": {
+                        "description": "关闭思考模式",
+                        "type": "bool",
+                        "hint": "关闭 Ollama 思考模式。",
+                    },
                     "custom_extra_body": {
                         "description": "自定义请求体参数",
                         "type": "dict",
@@ -2315,11 +2510,46 @@ CONFIG_METADATA_2 = {
                         "type": "int",
                         "hint": "超时时间，单位为秒。",
                     },
+                    "mimo-stt-system-prompt": {
+                        "description": "系统提示词",
+                        "type": "string",
+                        "hint": "用于指导 MiMo STT 转录行为的 system prompt。",
+                    },
+                    "mimo-stt-user-prompt": {
+                        "description": "用户提示词",
+                        "type": "string",
+                        "hint": "附加给 MiMo STT 的用户提示词，用于约束返回结果格式。",
+                    },
                     "openai-tts-voice": {
                         "description": "voice",
                         "type": "string",
                         "hint": "OpenAI TTS 的声音。OpenAI 默认支持：'alloy', 'echo', 'fable', 'onyx', 'nova', 'shimmer'",
                     },
+                    "mimo-tts-voice": {
+                        "description": "音色",
+                        "type": "string",
+                        "hint": "MiMo TTS 的音色名称。可选值包括 'mimo_default'、'default_en'、'default_zh'。",
+                    },
+                    "mimo-tts-format": {
+                        "description": "输出格式",
+                        "type": "string",
+                        "hint": "MiMo TTS 生成音频的格式。支持 'wav'、'mp3'、'pcm'。",
+                    },
+                    "mimo-tts-style-prompt": {
+                        "description": "风格提示词",
+                        "type": "string",
+                        "hint": "会以 <style>...</style> 标签形式添加到待合成文本开头，用于控制语速、情绪、角色或风格，例如 开心、变快、孙悟空、悄悄话。可留空。",
+                    },
+                    "mimo-tts-dialect": {
+                        "description": "方言",
+                        "type": "string",
+                        "hint": "会与风格提示词一起写入开头的 <style>...</style> 标签中，例如 东北话、四川话、河南话、粤语。可留空。",
+                    },
+                    "mimo-tts-seed-text": {
+                        "description": "种子文本",
+                        "type": "string",
+                        "hint": "作为可选的 user 消息发送，用于辅助调节语气和风格，不会拼接到待合成文本中。",
+                    },
                     "fishaudio-tts-character": {
                         "description": "character",
                         "type": "string",
@@ -2335,6 +2565,12 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "hint": "启用前请 pip 安装 openai-whisper 库（N卡用户大约下载 2GB，主要是 torch 和 cuda，CPU 用户大约下载 1 GB），并且安装 ffmpeg。否则将无法正常转文字。",
                     },
+                    "whisper_device": {
+                        "description": "推理设备",
+                        "type": "string",
+                        "hint": "Whisper 推理设备。Apple Silicon 可选 mps；其他环境建议使用 cpu。若指定 mps 但当前环境不可用，将自动回退到 cpu。",
+                        "options": ["cpu", "mps"],
+                    },
                     "id": {
                         "description": "ID",
                         "type": "string",
@@ -2437,12 +2673,12 @@ CONFIG_METADATA_2 = {
                     "deerflow_assistant_id": {
                         "description": "Assistant ID",
                         "type": "string",
-                        "hint": "LangGraph assistant_id，默认为 lead_agent。",
+                        "hint": "DeerFlow 2.0 LangGraph assistant_id，默认为 lead_agent。",
                     },
                     "deerflow_model_name": {
                         "description": "模型名称覆盖",
                         "type": "string",
-                        "hint": "可选。覆盖 DeerFlow 默认模型（对应 runtime context 的 model_name）。",
+                        "hint": "可选。覆盖 DeerFlow 默认模型（对应运行时 configurable 的 model_name）。",
                     },
                     "deerflow_thinking_enabled": {
                         "description": "启用思考模式",
@@ -2451,17 +2687,17 @@ CONFIG_METADATA_2 = {
                     "deerflow_plan_mode": {
                         "description": "启用计划模式",
                         "type": "bool",
-                        "hint": "对应 DeerFlow 的 is_plan_mode。",
+                        "hint": "对应 DeerFlow 2.0 运行时 configurable 的 is_plan_mode。",
                     },
                     "deerflow_subagent_enabled": {
                         "description": "启用子智能体",
                         "type": "bool",
-                        "hint": "对应 DeerFlow 的 subagent_enabled。",
+                        "hint": "对应 DeerFlow 2.0 运行时 configurable 的 subagent_enabled。",
                     },
                     "deerflow_max_concurrent_subagents": {
                         "description": "子智能体最大并发数",
                         "type": "int",
-                        "hint": "对应 DeerFlow 的 max_concurrent_subagents。仅在启用子智能体时生效，默认 3。",
+                        "hint": "对应 DeerFlow 2.0 运行时 configurable 的 max_concurrent_subagents。仅在启用子智能体时生效，默认 3。",
                     },
                     "deerflow_recursion_limit": {
                         "description": "递归深度上限",
@@ -2946,7 +3182,12 @@ CONFIG_METADATA_3 = {
                     "provider_settings.websearch_provider": {
                         "description": "网页搜索提供商",
                         "type": "string",
-                        "options": ["default", "tavily", "baidu_ai_search", "bocha"],
+                        "options": [
+                            "tavily",
+                            "baidu_ai_search",
+                            "bocha",
+                            "brave",
+                        ],
                         "condition": {
                             "provider_settings.web_search": True,
                         },
@@ -2971,6 +3212,16 @@ CONFIG_METADATA_3 = {
                             "provider_settings.web_search": True,
                         },
                     },
+                    "provider_settings.websearch_brave_key": {
+                        "description": "Brave Search API Key",
+                        "type": "list",
+                        "items": {"type": "string"},
+                        "hint": "可添加多个 Key 进行轮询。",
+                        "condition": {
+                            "provider_settings.websearch_provider": "brave",
+                            "provider_settings.web_search": True,
+                        },
+                    },
                     "provider_settings.websearch_baidu_app_builder_key": {
                         "description": "百度千帆智能云 APP Builder API Key",
                         "type": "string",
@@ -3323,14 +3574,39 @@ CONFIG_METADATA_3 = {
                         "type": "string",
                         "hint": "可使用 {{prompt}} 作为用户输入的占位符。如果不输入占位符则代表添加在用户输入的前面。",
                     },
+                    "provider_settings.image_compress_enabled": {
+                        "description": "启用图片压缩",
+                        "type": "bool",
+                        "hint": "启用后，发送给多模态模型前会先压缩本地大图片。",
+                    },
+                    "provider_settings.image_compress_options.max_size": {
+                        "description": "最大边长",
+                        "type": "int",
+                        "hint": "压缩后图片的最长边，单位为像素。超过该尺寸时会按比例缩放。",
+                        "condition": {
+                            "provider_settings.image_compress_enabled": True,
+                        },
+                        "slider": {"min": 256, "max": 4096, "step": 64},
+                    },
+                    "provider_settings.image_compress_options.quality": {
+                        "description": "压缩质量",
+                        "type": "int",
+                        "hint": "JPEG 输出质量，范围为 1-100。值越高，画质越好，文件也越大。",
+                        "condition": {
+                            "provider_settings.image_compress_enabled": True,
+                        },
+                        "slider": {"min": 1, "max": 100, "step": 1},
+                    },
                     "provider_tts_settings.dual_output": {
                         "description": "开启 TTS 时同时输出语音和文字内容",
                         "type": "bool",
+                        "collapsed": True,
                     },
                     "provider_settings.reachability_check": {
                         "description": "提供商可达性检测",
                         "type": "bool",
                         "hint": "/provider 命令列出模型时是否并发检测连通性。开启后会主动调用模型测试连通性，可能产生额外 token 消耗。",
+                        "collapsed": True,
                     },
                     "provider_settings.max_quoted_fallback_images": {
                         "description": "引用图片回退解析上限",
@@ -3339,6 +3615,7 @@ CONFIG_METADATA_3 = {
                         "condition": {
                             "provider_settings.agent_runner_type": "local",
                         },
+                        "collapsed": True,
                     },
                     "provider_settings.quoted_message_parser.max_component_chain_depth": {
                         "description": "引用解析组件链深度",
@@ -3347,6 +3624,7 @@ CONFIG_METADATA_3 = {
                         "condition": {
                             "provider_settings.agent_runner_type": "local",
                         },
+                        "collapsed": True,
                     },
                     "provider_settings.quoted_message_parser.max_forward_node_depth": {
                         "description": "引用解析转发节点深度",
@@ -3355,6 +3633,7 @@ CONFIG_METADATA_3 = {
                         "condition": {
                             "provider_settings.agent_runner_type": "local",
                         },
+                        "collapsed": True,
                     },
                     "provider_settings.quoted_message_parser.max_forward_fetch": {
                         "description": "引用解析转发拉取上限",
@@ -3363,6 +3642,7 @@ CONFIG_METADATA_3 = {
                         "condition": {
                             "provider_settings.agent_runner_type": "local",
                         },
+                        "collapsed": True,
                     },
                     "provider_settings.quoted_message_parser.warn_on_action_failure": {
                         "description": "引用解析 action 失败告警",
@@ -3371,6 +3651,7 @@ CONFIG_METADATA_3 = {
                         "condition": {
                             "provider_settings.agent_runner_type": "local",
                         },
+                        "collapsed": True,
                     },
                 },
                 "condition": {
@@ -3445,7 +3726,7 @@ CONFIG_METADATA_3 = {
                         "description": "白名单 ID 列表",
                         "type": "list",
                         "items": {"type": "string"},
-                        "hint": "使用 /sid 获取 ID。",
+                        "hint": "使用 /sid 获取 ID。当白名单列表为空时，代表不启用白名单（即所有 ID 都在白名单内）。",
                     },
                     "platform_settings.id_whitelist_log": {
                         "description": "输出日志",
@@ -3758,6 +4039,13 @@ CONFIG_METADATA_3_SYSTEM = {
                 "description": "系统配置",
                 "type": "object",
                 "items": {
+                    "language": {
+                        "description": "系统语言",
+                        "type": "string",
+                        "hint": "用于 AstrBot 运行时回复的语言。目前支持简体中文和英文。",
+                        "options": [Language.ZH_CN.value, Language.EN_US.value],
+                        "labels": ["简体中文", "English"],
+                    },
                     "t2i_strategy": {
                         "description": "文本转图像策略",
                         "type": "string",
@@ -3872,9 +4160,9 @@ CONFIG_METADATA_3_SYSTEM = {
                         "hint": "时区设置。请填写 IANA 时区名称, 如 Asia/Shanghai, 为空时使用系统默认时区。所有时区请查看: https://data.iana.org/time-zones/tzdb-2021a/zone1970.tab",
                     },
                     "http_proxy": {
-                        "description": "HTTP 代理",
+                        "description": "代理",
                         "type": "string",
-                        "hint": "启用后，会以添加环境变量的方式设置代理。格式为 `http://ip:port`",
+                        "hint": "启用后，会以添加环境变量的方式设置代理。支持 http://、https://、socks5:// 格式，例如：http://127.0.0.1:7890 或 socks5://127.0.0.1:7891",
                     },
                     "no_proxy": {
                         "description": "直连地址列表",

astrbot/core/cron/manager.py

@@ -275,8 +275,8 @@ class CronJobManager:
         )
         from astrbot.core.astr_main_agent_resources import (
             PROACTIVE_AGENT_CRON_WOKE_SYSTEM_PROMPT,
-            SEND_MESSAGE_TO_USER_TOOL,
         )
+        from astrbot.core.tools.message_tools import SendMessageToUserTool
 
         try:
             session = (
@@ -307,8 +307,11 @@ class CronJobManager:
         if cron_payload.get("origin", "tool") == "api":
             cron_event.role = "admin"
 
+        tool_call_timeout = cfg.get("provider_settings", {}).get(
+            "tool_call_timeout", 120
+        )
         config = MainAgentBuildConfig(
-            tool_call_timeout=3600,
+            tool_call_timeout=tool_call_timeout,
             llm_safety_mode=False,
             streaming_response=False,
         )
@@ -339,7 +342,9 @@ class CronJobManager:
         )
         if not req.func_tool:
             req.func_tool = ToolSet()
-        req.func_tool.add_tool(SEND_MESSAGE_TO_USER_TOOL)
+        req.func_tool.add_tool(
+            self.ctx.get_llm_tool_manager().get_builtin_tool(SendMessageToUserTool)
+        )
 
         result = await build_main_agent(
             event=cron_event, plugin_context=self.ctx, config=config, req=req

astrbot/core/db/__init__.py

@@ -21,6 +21,7 @@ from astrbot.core.db.po import (
     PlatformSession,
     PlatformStat,
     Preference,
+    ProviderStat,
     SessionProjectRelation,
     Stats,
 )
@@ -33,10 +34,18 @@ class BaseDatabase(abc.ABC):
     DATABASE_URL = ""
 
     def __init__(self) -> None:
+        # SQLite only supports a single writer at a time.  Without a busy
+        # timeout the driver raises "database is locked" instantly when a
+        # second write is attempted.  Setting timeout=30 tells SQLite to
+        # wait up to 30 s for the lock, which is enough to ride out brief
+        # write bursts from concurrent agent/metrics/session operations.
+        is_sqlite = "sqlite" in self.DATABASE_URL
+        connect_args = {"timeout": 30} if is_sqlite else {}
         self.engine = create_async_engine(
             self.DATABASE_URL,
             echo=False,
             future=True,
+            connect_args=connect_args,
         )
         self.AsyncSessionLocal = async_sessionmaker(
             self.engine,
@@ -97,6 +106,21 @@ class BaseDatabase(abc.ABC):
         """Get platform statistics within the specified offset in seconds and group by platform_id."""
         ...
 
+    @abc.abstractmethod
+    async def insert_provider_stat(
+        self,
+        *,
+        umo: str,
+        provider_id: str,
+        provider_model: str | None = None,
+        conversation_id: str | None = None,
+        status: str = "completed",
+        stats: dict | None = None,
+        agent_type: str = "internal",
+    ) -> ProviderStat:
+        """Insert a per-response provider stat record."""
+        ...
+
     @abc.abstractmethod
     async def get_conversations(
         self,

astrbot/core/db/po.py

@@ -38,6 +38,30 @@ class PlatformStat(SQLModel, table=True):
     )
 
 
+class ProviderStat(TimestampMixin, SQLModel, table=True):
+    """Per-response provider stats for internal agent runs."""
+
+    __tablename__: str = "provider_stats"
+
+    id: int | None = Field(
+        default=None,
+        primary_key=True,
+        sa_column_kwargs={"autoincrement": True},
+    )
+    agent_type: str = Field(default="internal", nullable=False, index=True)
+    status: str = Field(default="completed", nullable=False, index=True)
+    umo: str = Field(nullable=False, index=True)
+    conversation_id: str | None = Field(default=None, index=True)
+    provider_id: str = Field(nullable=False, index=True)
+    provider_model: str | None = Field(default=None, index=True)
+    token_input_other: int = Field(default=0, nullable=False)
+    token_input_cached: int = Field(default=0, nullable=False)
+    token_output: int = Field(default=0, nullable=False)
+    start_time: float = Field(default=0.0, nullable=False)
+    end_time: float = Field(default=0.0, nullable=False)
+    time_to_first_token: float = Field(default=0.0, nullable=False)
+
+
 class ConversationV2(TimestampMixin, SQLModel, table=True):
     __tablename__: str = "conversations"
 

astrbot/core/db/sqlite.py

@@ -23,6 +23,7 @@ from astrbot.core.db.po import (
     PlatformSession,
     PlatformStat,
     Preference,
+    ProviderStat,
     SessionProjectRelation,
     SQLModel,
 )
@@ -169,6 +170,51 @@ class SQLiteDatabase(BaseDatabase):
             )
             return list(result.scalars().all())
 
+    async def insert_provider_stat(
+        self,
+        *,
+        umo: str,
+        provider_id: str,
+        provider_model: str | None = None,
+        conversation_id: str | None = None,
+        status: str = "completed",
+        stats: dict | None = None,
+        agent_type: str = "internal",
+    ) -> ProviderStat:
+        """Insert a provider stat record for a single agent response."""
+        stats = stats or {}
+        token_usage = stats.get("token_usage", {})
+
+        token_input_other = int(token_usage.get("input_other", 0) or 0)
+        token_input_cached = int(token_usage.get("input_cached", 0) or 0)
+        token_output = int(token_usage.get("output", 0) or 0)
+
+        start_time = float(stats.get("start_time", 0.0) or 0.0)
+        end_time = float(stats.get("end_time", 0.0) or 0.0)
+        time_to_first_token = float(stats.get("time_to_first_token", 0.0) or 0.0)
+
+        async with self.get_db() as session:
+            session: AsyncSession
+            async with session.begin():
+                record = ProviderStat(
+                    agent_type=agent_type,
+                    status=status,
+                    umo=umo,
+                    conversation_id=conversation_id,
+                    provider_id=provider_id,
+                    provider_model=provider_model,
+                    token_input_other=token_input_other,
+                    token_input_cached=token_input_cached,
+                    token_output=token_output,
+                    start_time=start_time,
+                    end_time=end_time,
+                    time_to_first_token=time_to_first_token,
+                )
+                session.add(record)
+                await session.flush()
+                await session.refresh(record)
+                return record
+
     # ====
     # Conversation Management
     # ====

astrbot/core/db/vec_db/faiss_impl/vec_db.py

@@ -74,6 +74,12 @@ class FaissVecDB(BaseVecDB):
         metadatas = metadatas or [{} for _ in contents]
         ids = ids or [str(uuid.uuid4()) for _ in contents]
 
+        if not contents:
+            logger.debug(
+                "No contents provided for batch insert; skipping embedding generation."
+            )
+            return []
+
         start = time.time()
         logger.debug(f"Generating embeddings for {len(contents)} contents...")
         vectors = await self.embedding_provider.get_embeddings_batch(

astrbot/core/exceptions.py

@@ -7,3 +7,7 @@ class AstrBotError(Exception):
 
 class ProviderNotFoundError(AstrBotError):
     """Raised when a specified provider is not found."""
+
+
+class EmptyModelOutputError(AstrBotError):
+    """Raised when the model response contains no usable assistant output."""

astrbot/core/i18n/__init__.py

@@ -0,0 +1,62 @@
+import json
+from enum import Enum
+from functools import lru_cache
+from pathlib import Path
+from typing import Any
+
+CORE_LOCALE_DIR = Path(__file__).resolve().parent / "locales"
+
+
+class Language(str, Enum):
+    ZH_CN = "zh-CN"
+    EN_US = "en-US"
+
+
+DEFAULT_LANGUAGE = Language.ZH_CN.value
+
+
+def normalize_language(language: str | Language | None) -> str:
+    if isinstance(language, Language):
+        return language.value
+    if language == Language.EN_US.value:
+        return Language.EN_US.value
+    return Language.ZH_CN.value
+
+
+@lru_cache(maxsize=64)
+def _load_locale(locale_dir: str, language: str) -> dict[str, Any]:
+    locale_path = Path(locale_dir) / f"{language}.json"
+    with locale_path.open(encoding="utf-8") as f:
+        return json.load(f)
+
+
+def _resolve_key(data: dict[str, Any], key: str) -> Any:
+    value: Any = data
+    for part in key.split("."):
+        if not isinstance(value, dict) or part not in value:
+            return None
+        value = value[part]
+    return value
+
+
+def t(
+    translation_key: str,
+    *,
+    locale: str | None = None,
+    locale_dir: str | Path | None = None,
+    **kwargs: Any,
+) -> str:
+    language = normalize_language(locale)
+    resolved_locale_dir = str(locale_dir or CORE_LOCALE_DIR)
+    text = _resolve_key(_load_locale(resolved_locale_dir, language), translation_key)
+
+    if text is None and language != DEFAULT_LANGUAGE:
+        text = _resolve_key(
+            _load_locale(resolved_locale_dir, DEFAULT_LANGUAGE),
+            translation_key,
+        )
+    if not isinstance(text, str):
+        return translation_key
+    if not kwargs:
+        return text
+    return text.format(**kwargs)

astrbot/core/i18n/locales/en-US.json

@@ -0,0 +1,12 @@
+{
+  "pipeline": {
+    "filter_error": "Plugin {plugin_name}: {error}",
+    "no_permission": "You (ID: {sender_id}) do not have permission to use this command. Use /sid to get your ID and ask an administrator to add it.",
+    "content_blocked": "Your message or the model response contains inappropriate content and has been blocked.",
+    "keyword_blocked_reason": "Content safety check failed because a sensitive keyword was matched.",
+    "baidu_aip_violation_header": "Baidu content moderation found {count} violations:\n",
+    "baidu_aip_conclusion": "\nConclusion: {conclusion}",
+    "plugin_handler_error": ":(\n\nAn exception occurred while calling plugin {plugin_name}'s handler {handler_name}: {error}",
+    "reasoning_prefix": "🤔 Thinking: {reasoning_content}\n"
+  }
+}

astrbot/core/i18n/locales/zh-CN.json

@@ -0,0 +1,12 @@
+{
+  "pipeline": {
+    "filter_error": "插件 {plugin_name}: {error}",
+    "no_permission": "您(ID: {sender_id})的权限不足以使用此指令。通过 /sid 获取 ID 并请管理员添加。",
+    "content_blocked": "你的消息或者大模型的响应中包含不适当的内容，已被屏蔽。",
+    "keyword_blocked_reason": "内容安全检查不通过，匹配到敏感词。",
+    "baidu_aip_violation_header": "百度审核服务发现 {count} 处违规：\n",
+    "baidu_aip_conclusion": "\n判断结果：{conclusion}",
+    "plugin_handler_error": ":(\n\n在调用插件 {plugin_name} 的处理函数 {handler_name} 时出现异常：{error}",
+    "reasoning_prefix": "🤔 思考: {reasoning_content}\n"
+  }
+}

astrbot/core/knowledge_base/kb_db_sqlite.py

@@ -1,12 +1,12 @@
 from contextlib import asynccontextmanager
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 from sqlalchemy import delete, func, select, text, update
 from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker, create_async_engine
 from sqlmodel import col, desc
 
 from astrbot.core import logger
-from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
 from astrbot.core.knowledge_base.models import (
     BaseKBModel,
     KBDocument,
@@ -15,6 +15,9 @@ from astrbot.core.knowledge_base.models import (
 )
 from astrbot.core.utils.astrbot_path import get_astrbot_knowledge_base_path
 
+if TYPE_CHECKING:
+    from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
+
 
 class KBSQLiteDatabase:
     def __init__(self, db_path: str | None = None) -> None:
@@ -296,7 +299,7 @@ class KBSQLiteDatabase:
 
         return metadata_map
 
-    async def delete_document_by_id(self, doc_id: str, vec_db: FaissVecDB) -> None:
+    async def delete_document_by_id(self, doc_id: str, vec_db: "FaissVecDB") -> None:
         """删除单个文档及其相关数据"""
         # 在知识库表中删除
         async with self.get_db() as session, session.begin():
@@ -324,7 +327,7 @@ class KBSQLiteDatabase:
             result = await session.execute(stmt)
             return result.scalar_one_or_none()
 
-    async def update_kb_stats(self, kb_id: str, vec_db: FaissVecDB) -> None:
+    async def update_kb_stats(self, kb_id: str, vec_db: "FaissVecDB") -> None:
         """更新知识库统计信息"""
         chunk_cnt = await vec_db.count_documents()
 

astrbot/core/knowledge_base/kb_helper.py

@@ -4,12 +4,12 @@ import re
 import time
 import uuid
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 import aiofiles
 
 from astrbot.core import logger
 from astrbot.core.db.vec_db.base import BaseVecDB
-from astrbot.core.db.vec_db.faiss_impl.vec_db import FaissVecDB
 from astrbot.core.provider.manager import ProviderManager
 from astrbot.core.provider.provider import (
     EmbeddingProvider,
@@ -27,6 +27,9 @@ from .parsers.url_parser import extract_text_from_url
 from .parsers.util import select_parser
 from .prompts import TEXT_REPAIR_SYSTEM_PROMPT
 
+if TYPE_CHECKING:
+    from astrbot.core.db.vec_db.faiss_impl.vec_db import FaissVecDB
+
 
 class RateLimiter:
     """一个简单的速率限制器"""
@@ -108,6 +111,7 @@ Text chunk to process:
 class KBHelper:
     vec_db: BaseVecDB
     kb: KnowledgeBase
+    init_error: str | None
 
     def __init__(
         self,
@@ -122,6 +126,7 @@ class KBHelper:
         self.prov_mgr = provider_manager
         self.kb_root_dir = kb_root_dir
         self.chunker = chunker
+        self.init_error = None
 
         self.kb_dir = Path(self.kb_root_dir) / self.kb.kb_id
         self.kb_medias_dir = Path(self.kb_dir) / "medias" / self.kb.kb_id
@@ -148,21 +153,30 @@ class KBHelper:
     async def get_rp(self) -> RerankProvider | None:
         if not self.kb.rerank_provider_id:
             return None
-        rp: RerankProvider = await self.prov_mgr.get_provider_by_id(
+        rp: RerankProvider | None = await self.prov_mgr.get_provider_by_id(
             self.kb.rerank_provider_id,
         )  # type: ignore
         if not rp:
-            raise ValueError(
-                f"无法找到 ID 为 {self.kb.rerank_provider_id} 的 Rerank Provider",
+            logger.warning(
+                f"知识库 {self.kb.kb_name}({self.kb.kb_id}) 的 Rerank Provider({self.kb.rerank_provider_id}) 不可用，将跳过重排序。",
             )
+            return None
         return rp
 
-    async def _ensure_vec_db(self) -> FaissVecDB:
+    async def _ensure_vec_db(self) -> "FaissVecDB":
         if not self.kb.embedding_provider_id:
             raise ValueError(f"知识库 {self.kb.kb_name} 未配置 Embedding Provider")
 
         ep = await self.get_ep()
-        rp = await self.get_rp()
+        rp: RerankProvider | None = None
+        try:
+            rp = await self.get_rp()
+        except Exception as e:
+            logger.warning(
+                f"知识库 {self.kb.kb_name}({self.kb.kb_id}) 初始化重排序能力失败，将跳过重排序: {e}",
+            )
+
+        from astrbot.core.db.vec_db.faiss_impl.vec_db import FaissVecDB
 
         vec_db = FaissVecDB(
             doc_store_path=str(self.kb_dir / "doc.db"),
@@ -172,6 +186,8 @@ class KBHelper:
         )
         await vec_db.initialize()
         self.vec_db = vec_db
+        # Clear stale init_error once initialization succeeds.
+        self.init_error = None
         return vec_db
 
     async def delete_vec_db(self) -> None:
@@ -183,7 +199,7 @@ class KBHelper:
             shutil.rmtree(self.kb_dir)
 
     async def terminate(self) -> None:
-        if self.vec_db:
+        if hasattr(self, "vec_db") and self.vec_db:
             await self.vec_db.close()
 
     async def upload_document(

astrbot/core/knowledge_base/kb_mgr.py

@@ -1,4 +1,3 @@
-import traceback
 from pathlib import Path
 
 from astrbot.core import logger
@@ -56,8 +55,7 @@ class KnowledgeBaseManager:
             logger.error(f"知识库模块导入失败: {e}")
             logger.warning("请确保已安装所需依赖: pypdf, aiofiles, Pillow, rank-bm25")
         except Exception as e:
-            logger.error(f"知识库模块初始化失败: {e}")
-            logger.error(traceback.format_exc())
+            logger.error(f"知识库模块初始化失败: {e}", exc_info=True)
 
     async def _init_kb_database(self) -> None:
         self.kb_db = KBSQLiteDatabase(DB_PATH.as_posix())
@@ -76,7 +74,14 @@ class KnowledgeBaseManager:
                 kb_root_dir=FILES_PATH,
                 chunker=CHUNKER,
             )
-            await kb_helper.initialize()
+            try:
+                await kb_helper.initialize()
+            except Exception as e:
+                kb_helper.init_error = str(e)
+                logger.error(
+                    f"知识库 {record.kb_name}({record.kb_id}) 初始化失败: {e}",
+                    exc_info=True,
+                )
             self.kb_insts[record.kb_id] = kb_helper
 
     async def create_kb(
@@ -179,6 +184,20 @@ class KnowledgeBaseManager:
             return None
 
         kb = kb_helper.kb
+        previous_state = {
+            "kb_name": kb.kb_name,
+            "description": kb.description,
+            "emoji": kb.emoji,
+            "embedding_provider_id": kb.embedding_provider_id,
+            "rerank_provider_id": kb.rerank_provider_id,
+            "chunk_size": kb.chunk_size,
+            "chunk_overlap": kb.chunk_overlap,
+            "top_k_dense": kb.top_k_dense,
+            "top_k_sparse": kb.top_k_sparse,
+            "top_m_final": kb.top_m_final,
+        }
+        previous_init_error = kb_helper.init_error
+
         if kb_name is not None:
             kb.kb_name = kb_name
         if description is not None:
@@ -198,12 +217,47 @@ class KnowledgeBaseManager:
             kb.top_k_sparse = top_k_sparse
         if top_m_final is not None:
             kb.top_m_final = top_m_final
+
+        # Build a new helper first. Keep current vec_db alive until new init succeeds.
+        new_helper = KBHelper(
+            kb_db=self.kb_db,
+            kb=kb,
+            provider_manager=self.provider_manager,
+            kb_root_dir=FILES_PATH,
+            chunker=CHUNKER,
+        )
+
+        try:
+            await new_helper.initialize()
+        except Exception as e:
+            # Roll back in-memory settings and keep current helper available.
+            kb.kb_name = previous_state["kb_name"]
+            kb.description = previous_state["description"]
+            kb.emoji = previous_state["emoji"]
+            kb.embedding_provider_id = previous_state["embedding_provider_id"]
+            kb.rerank_provider_id = previous_state["rerank_provider_id"]
+            kb.chunk_size = previous_state["chunk_size"]
+            kb.chunk_overlap = previous_state["chunk_overlap"]
+            kb.top_k_dense = previous_state["top_k_dense"]
+            kb.top_k_sparse = previous_state["top_k_sparse"]
+            kb.top_m_final = previous_state["top_m_final"]
+            kb_helper.init_error = previous_init_error
+            logger.error(
+                f"知识库 {kb.kb_name}({kb.kb_id}) 重新初始化失败，继续使用旧实例: {e}",
+                exc_info=True,
+            )
+            return kb_helper
+
         async with self.kb_db.get_db() as session:
             session.add(kb)
             await session.commit()
             await session.refresh(kb)
 
-        return kb_helper
+        old_helper = kb_helper
+        self.kb_insts[kb_id] = new_helper
+        await old_helper.terminate()
+        new_helper.init_error = None
+        return new_helper
 
     async def retrieve(
         self,
@@ -215,11 +269,21 @@ class KnowledgeBaseManager:
         """从指定知识库中检索相关内容"""
         kb_ids = []
         kb_id_helper_map = {}
+        unavailable_kbs = []
         for kb_name in kb_names:
             if kb_helper := await self.get_kb_by_name(kb_name):
+                if kb_helper.init_error:
+                    unavailable_kbs.append((kb_name, kb_helper.init_error))
+                    logger.warning(f"知识库 {kb_name} 不可用: {kb_helper.init_error}")
+                    continue
                 kb_ids.append(kb_helper.kb.kb_id)
                 kb_id_helper_map[kb_helper.kb.kb_id] = kb_helper
 
+        # all requested KBs are unavailable
+        if not kb_ids and unavailable_kbs:
+            errors = "; ".join(f"{n}: {e}" for n, e in unavailable_kbs)
+            raise ValueError(f"所有请求的知识库均不可用: {errors}")
+
         if not kb_ids:
             return {}
 

astrbot/core/knowledge_base/retrieval/manager.py

@@ -5,10 +5,10 @@
 
 import time
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
 
 from astrbot import logger
 from astrbot.core.db.vec_db.base import Result
-from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
 from astrbot.core.knowledge_base.kb_db_sqlite import KBSQLiteDatabase
 from astrbot.core.knowledge_base.retrieval.rank_fusion import RankFusion
 from astrbot.core.knowledge_base.retrieval.sparse_retriever import SparseRetriever
@@ -16,6 +16,9 @@ from astrbot.core.provider.provider import RerankProvider
 
 from ..kb_helper import KBHelper
 
+if TYPE_CHECKING:
+    from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
+
 
 @dataclass
 class RetrievalResult:
@@ -170,26 +173,31 @@ class RetrievalManager:
         first_rerank = None
         for kb_id in kb_ids:
             vec_db = kb_options[kb_id]["vec_db"]
-            if not isinstance(vec_db, FaissVecDB):
-                logger.warning(f"vec_db for kb_id {kb_id} is not FaissVecDB")
+            rerank_provider = (
+                getattr(vec_db, "rerank_provider", None) if vec_db else None
+            )
+            if rerank_provider is None:
                 continue
 
             rerank_pi = kb_options[kb_id]["rerank_provider_id"]
             if (
                 vec_db
-                and vec_db.rerank_provider
+                and rerank_provider
                 and rerank_pi
-                and rerank_pi == vec_db.rerank_provider.meta().id
+                and rerank_pi == rerank_provider.meta().id
             ):
-                first_rerank = vec_db.rerank_provider
+                first_rerank = rerank_provider
                 break
         if first_rerank and retrieval_results:
-            retrieval_results = await self._rerank(
-                query=query,
-                results=retrieval_results,
-                top_k=top_m_final,
-                rerank_provider=first_rerank,
-            )
+            try:
+                retrieval_results = await self._rerank(
+                    query=query,
+                    results=retrieval_results,
+                    top_k=top_m_final,
+                    rerank_provider=first_rerank,
+                )
+            except Exception as e:
+                logger.warning(f"Rerank 执行失败，已跳过重排序并使用融合结果: {e}")
 
         return retrieval_results[:top_m_final]
 
@@ -229,10 +237,10 @@ class RetrievalManager:
 
                 all_results.extend(vec_results)
             except Exception as e:
-                from astrbot.core import logger
-
-                logger.warning(f"知识库 {kb_id} 稠密检索失败: {e}")
-                continue
+                logger.error(f"知识库 {kb_id} 稠密检索失败: {e}", exc_info=True)
+                if len(kb_ids) == 1:
+                    raise RuntimeError(f"知识库 {kb_id} 稠密检索失败: {e}") from e
+                # multi-KB: skip the faulty KB and continue
 
         # 按相似度排序并返回 top_k
         all_results.sort(key=lambda x: x.similarity, reverse=True)

astrbot/core/knowledge_base/retrieval/sparse_retriever.py

@@ -6,13 +6,16 @@
 import json
 import os
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
 
 import jieba
 from rank_bm25 import BM25Okapi
 
-from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
 from astrbot.core.knowledge_base.kb_db_sqlite import KBSQLiteDatabase
 
+if TYPE_CHECKING:
+    from astrbot.core.db.vec_db.faiss_impl import FaissVecDB
+
 
 @dataclass
 class SparseResult:
@@ -73,7 +76,7 @@ class SparseRetriever:
         top_k_sparse = 0
         chunks = []
         for kb_id in kb_ids:
-            vec_db: FaissVecDB = kb_options.get(kb_id, {}).get("vec_db")
+            vec_db: FaissVecDB | None = kb_options.get(kb_id, {}).get("vec_db")
             if not vec_db:
                 continue
             result = await vec_db.document_storage.get_documents(

astrbot/core/message/components.py

@@ -64,7 +64,6 @@ class ComponentType(str, Enum):
     Music = "Music"
     Json = "Json"
     Unknown = "Unknown"
-    WechatEmoji = "WechatEmoji"  # Wechat 下的 emoji 表情包
 
 
 class BaseMessageComponent(BaseModel):
@@ -91,7 +90,6 @@ class BaseMessageComponent(BaseModel):
 class Plain(BaseMessageComponent):
     type: ComponentType = ComponentType.Plain
     text: str
-    convert: bool | None = True
 
     def __init__(self, text: str, convert: bool = True, **_) -> None:
         super().__init__(text=text, convert=convert, **_)
@@ -114,15 +112,11 @@ class Face(BaseMessageComponent):
 class Record(BaseMessageComponent):
     type: ComponentType = ComponentType.Record
     file: str | None = ""
-    magic: bool | None = False
     url: str | None = ""
-    cache: bool | None = True
-    proxy: bool | None = True
-    timeout: int | None = 0
     # Original text content (e.g. TTS source text), used as caption in fallback scenarios
     text: str | None = None
     # 额外
-    path: str | None
+    path: str | None = None
 
     def __init__(self, file: str | None, **_) -> None:
         for k in _:
@@ -224,7 +218,6 @@ class Video(BaseMessageComponent):
     type: ComponentType = ComponentType.Video
     file: str
     cover: str | None = ""
-    c: int | None = 2
     # 额外
     path: str | None = ""
 
@@ -401,14 +394,9 @@ class Image(BaseMessageComponent):
     type: ComponentType = ComponentType.Image
     file: str | None = ""
     _type: str | None = ""
-    subType: int | None = 0
     url: str | None = ""
-    cache: bool | None = True
-    id: int | None = 40000
-    c: int | None = 2
     # 额外
     path: str | None = ""
-    file_unique: str | None = ""  # 某些平台可能有图片缓存的唯一标识
 
     def __init__(self, file: str | None, **_) -> None:
         super().__init__(file=file, **_)
@@ -839,16 +827,6 @@ class File(BaseMessageComponent):
         }
 
 
-class WechatEmoji(BaseMessageComponent):
-    type: ComponentType = ComponentType.WechatEmoji
-    md5: str | None = ""
-    md5_len: int | None = 0
-    cdnurl: str | None = ""
-
-    def __init__(self, **_) -> None:
-        super().__init__(**_)
-
-
 ComponentTypes = {
     # Basic Message Segments
     "plain": Plain,
@@ -874,5 +852,4 @@ ComponentTypes = {
     "nodes": Nodes,
     "json": Json,
     "unknown": Unknown,
-    "WechatEmoji": WechatEmoji,
 }

astrbot/core/persona_mgr.py

@@ -44,6 +44,22 @@ class PersonaManager:
             raise ValueError(f"Persona with ID {persona_id} does not exist.")
         return persona
 
+    def get_persona_v3_by_id(self, persona_id: str | None) -> Personality | None:
+        """Resolve a v3 persona object by id.
+
+        - None/empty id returns None.
+        - "default" maps to in-memory DEFAULT_PERSONALITY.
+        - Otherwise search in personas_v3 by persona name.
+        """
+        if not persona_id:
+            return None
+        if persona_id == "default":
+            return DEFAULT_PERSONALITY
+        return next(
+            (persona for persona in self.personas_v3 if persona["name"] == persona_id),
+            None,
+        )
+
     async def get_default_persona_v3(
         self,
         umo: str | MessageSession | None = None,
@@ -54,12 +70,7 @@ class PersonaManager:
             "default_personality",
             "default",
         )
-        if not default_persona_id or default_persona_id == "default":
-            return DEFAULT_PERSONALITY
-        try:
-            return next(p for p in self.personas_v3 if p["name"] == default_persona_id)
-        except Exception:
-            return DEFAULT_PERSONALITY
+        return self.get_persona_v3_by_id(default_persona_id) or DEFAULT_PERSONALITY
 
     async def resolve_selected_persona(
         self,

astrbot/core/pipeline/content_safety_check/stage.py

@@ -1,6 +1,7 @@
 from collections.abc import AsyncGenerator
 
 from astrbot.core import logger
+from astrbot.core.i18n import t
 from astrbot.core.message.message_event_result import MessageEventResult
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 
@@ -17,6 +18,7 @@ class ContentSafetyCheckStage(Stage):
     """
 
     async def initialize(self, ctx: PipelineContext) -> None:
+        self.ctx = ctx
         config = ctx.astrbot_config["content_safety"]
         self.strategy_selector = StrategySelector(config)
 
@@ -27,12 +29,13 @@ class ContentSafetyCheckStage(Stage):
     ) -> AsyncGenerator[None, None]:
         """检查内容安全"""
         text = check_text if check_text else event.get_message_str()
-        ok, info = self.strategy_selector.check(text)
+        locale = self.ctx.get_current_language()
+        ok, info = self.strategy_selector.check(text, locale=locale)
         if not ok:
             if event.is_at_or_wake_command:
                 event.set_result(
                     MessageEventResult().message(
-                        "你的消息或者大模型的响应中包含不适当的内容，已被屏蔽。",
+                        t("pipeline.content_blocked", locale=locale),
                     ),
                 )
                 yield

astrbot/core/pipeline/content_safety_check/strategies/__init__.py

@@ -3,5 +3,5 @@ import abc
 
 class ContentSafetyStrategy(abc.ABC):
     @abc.abstractmethod
-    def check(self, content: str) -> tuple[bool, str]:
+    def check(self, content: str, locale: str | None = None) -> tuple[bool, str]:
         raise NotImplementedError

astrbot/core/pipeline/content_safety_check/strategies/baidu_aip.py

@@ -4,6 +4,8 @@ from typing import Any, cast
 
 from aip import AipContentCensor
 
+from astrbot.core.i18n import t
+
 from . import ContentSafetyStrategy
 
 
@@ -14,7 +16,7 @@ class BaiduAipStrategy(ContentSafetyStrategy):
         self.secret_key = sk
         self.client = AipContentCensor(self.app_id, self.api_key, self.secret_key)
 
-    def check(self, content: str) -> tuple[bool, str]:
+    def check(self, content: str, locale: str | None = None) -> tuple[bool, str]:
         res = self.client.textCensorUserDefined(content)
         if "conclusionType" not in res:
             return False, ""
@@ -23,10 +25,18 @@ class BaiduAipStrategy(ContentSafetyStrategy):
         if "data" not in res:
             return False, ""
         count = len(res["data"])
-        parts = [f"百度审核服务发现 {count} 处违规：\n"]
+        parts = [
+            t("pipeline.baidu_aip_violation_header", locale=locale, count=count),
+        ]
         for i in res["data"]:
             # 百度 AIP 返回结构是动态 dict；类型检查时 i 可能被推断为序列，转成 dict 后用 get 取字段
             parts.append(f"{cast(dict[str, Any], i).get('msg', '')}；\n")
-        parts.append("\n判断结果：" + res["conclusion"])
+        parts.append(
+            t(
+                "pipeline.baidu_aip_conclusion",
+                locale=locale,
+                conclusion=res["conclusion"],
+            ),
+        )
         info = "".join(parts)
         return False, info

astrbot/core/pipeline/content_safety_check/strategies/keywords.py

@@ -1,5 +1,7 @@
 import re
 
+from astrbot.core.i18n import t
+
 from . import ContentSafetyStrategy
 
 
@@ -17,8 +19,8 @@ class KeywordsStrategy(ContentSafetyStrategy):
         #             json.loads(base64.b64decode(f.read()).decode("utf-8"))["keywords"]
         #         )
 
-    def check(self, content: str) -> tuple[bool, str]:
+    def check(self, content: str, locale: str | None = None) -> tuple[bool, str]:
         for keyword in self.keywords:
             if re.search(keyword, content):
-                return False, "内容安全检查不通过，匹配到敏感词。"
+                return False, t("pipeline.keyword_blocked_reason", locale=locale)
         return True, ""

astrbot/core/pipeline/content_safety_check/strategies/strategy.py

@@ -26,9 +26,9 @@ class StrategySelector:
                 ),
             )
 
-    def check(self, content: str) -> tuple[bool, str]:
+    def check(self, content: str, locale: str | None = None) -> tuple[bool, str]:
         for strategy in self.enabled_strategies:
-            ok, info = strategy.check(content)
+            ok, info = strategy.check(content, locale=locale)
             if not ok:
                 return False, info
         return True, ""

astrbot/core/pipeline/context.py

@@ -20,3 +20,6 @@ class PipelineContext:
     astrbot_config_id: str
     call_handler = call_handler
     call_event_hook = call_event_hook
+
+    def get_current_language(self) -> str:
+        return self.plugin_manager.context.get_current_language()

astrbot/core/pipeline/preprocess_stage/stage.py

@@ -6,6 +6,7 @@ from collections.abc import AsyncGenerator
 from astrbot.core import logger
 from astrbot.core.message.components import Image, Plain, Record
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from astrbot.core.utils.media_utils import ensure_wav
 
 from ..context import PipelineContext
 from ..stage import Stage, register_stage
@@ -64,6 +65,21 @@ class PreProcessStage(Stage):
                             logger.debug(f"路径映射: {url} -> {component.url}")
                     message_chain[idx] = component
 
+        # In here, we convert all Record components to wav format and update the file path.
+        message_chain = event.get_messages()
+        for idx, component in enumerate(message_chain):
+            if isinstance(component, Record):
+                try:
+                    original_path = await component.convert_to_file_path()
+                    record_path = await ensure_wav(original_path)
+                    if record_path != original_path:
+                        event.track_temporary_local_file(record_path)
+                    component.file = record_path
+                    component.path = record_path
+                    message_chain[idx] = component
+                except Exception as e:
+                    logger.warning(f"Voice processing failed: {e}")
+
         # STT
         if self.stt_settings.get("enable", False):
             # TODO: 独立
@@ -76,8 +92,8 @@ class PreProcessStage(Stage):
                 return
             message_chain = event.get_messages()
             for idx, component in enumerate(message_chain):
-                if isinstance(component, Record) and component.url:
-                    path = component.url.removeprefix("file://")
+                if isinstance(component, Record):
+                    path = await component.convert_to_file_path()
                     retry = 5
                     for i in range(retry):
                         try:

astrbot/core/pipeline/process_stage/follow_up.py

@@ -172,6 +172,9 @@ def try_capture_follow_up(event: AstrMessageEvent) -> FollowUpCapture | None:
     if not active_sender_id or active_sender_id != sender_id:
         return None
 
+    if runner_event.get_extra("agent_stop_requested"):
+        return None
+
     ticket = runner.follow_up(message_text=_event_follow_up_text(event))
     if not ticket:
         return None

astrbot/core/pipeline/process_stage/method/agent_sub_stages/internal.py

@@ -5,7 +5,7 @@ import base64
 from collections.abc import AsyncGenerator
 from dataclasses import replace
 
-from astrbot.core import logger
+from astrbot.core import db_helper, logger
 from astrbot.core.agent.message import Message
 from astrbot.core.agent.response import AgentStats
 from astrbot.core.astr_main_agent import (
@@ -13,7 +13,7 @@ from astrbot.core.astr_main_agent import (
     MainAgentBuildResult,
     build_main_agent,
 )
-from astrbot.core.message.components import File, Image
+from astrbot.core.message.components import File, Image, Record, Video
 from astrbot.core.message.message_event_result import (
     MessageChain,
     MessageEventResult,
@@ -144,6 +144,7 @@ class InternalAgentSubStage(Stage):
         follow_up_capture: FollowUpCapture | None = None
         follow_up_consumed_marked = False
         follow_up_activated = False
+        typing_requested = False
         try:
             streaming_response = self.streaming_response
             if (enable_streaming := event.get_extra("enable_streaming")) is not None:
@@ -152,7 +153,8 @@ class InternalAgentSubStage(Stage):
             has_provider_request = event.get_extra("provider_request") is not None
             has_valid_message = bool(event.message_str and event.message_str.strip())
             has_media_content = any(
-                isinstance(comp, Image | File) for comp in event.message_obj.message
+                isinstance(comp, (Image, File, Record, Video))
+                for comp in event.message_obj.message
             )
 
             if (
@@ -178,7 +180,11 @@ class InternalAgentSubStage(Stage):
                     )
                     return
 
-            await event.send_typing()
+            try:
+                typing_requested = True
+                await event.send_typing()
+            except Exception:
+                logger.warning("send_typing failed", exc_info=True)
             await call_event_hook(event, EventType.OnWaitingLLMRequestEvent)
 
             async with session_lock_manager.acquire_lock(event.unified_msg_origin):
@@ -345,6 +351,15 @@ class InternalAgentSubStage(Stage):
                         resp=final_resp.completion_text if final_resp else None,
                     )
 
+                    asyncio.create_task(
+                        _record_internal_agent_stats(
+                            event,
+                            req,
+                            agent_runner,
+                            final_resp,
+                        )
+                    )
+
                     # 检查事件是否被停止，如果被停止则不保存历史记录
                     if not event.is_stopped() or agent_runner.was_aborted():
                         await self._save_to_history(
@@ -377,6 +392,11 @@ class InternalAgentSubStage(Stage):
             )
             await event.send(MessageChain().message(error_text))
         finally:
+            if typing_requested:
+                try:
+                    await event.stop_typing()
+                except Exception:
+                    logger.warning("stop_typing failed", exc_info=True)
             if follow_up_capture:
                 await finalize_follow_up_capture(
                     follow_up_capture,
@@ -452,3 +472,46 @@ class InternalAgentSubStage(Stage):
 # these hosts are base64 encoded
 BLOCKED = {"dGZid2h2d3IuY2xvdWQuc2VhbG9zLmlv", "a291cmljaGF0"}
 decoded_blocked = [base64.b64decode(b).decode("utf-8") for b in BLOCKED]
+
+
+async def _record_internal_agent_stats(
+    event: AstrMessageEvent,
+    req: ProviderRequest | None,
+    agent_runner: AgentRunner | None,
+    final_resp: LLMResponse | None,
+) -> None:
+    """Persist internal agent stats without affecting the user response flow."""
+    if agent_runner is None:
+        return
+
+    provider = agent_runner.provider
+    stats = agent_runner.stats
+    if provider is None or stats is None:
+        return
+
+    try:
+        provider_config = getattr(provider, "provider_config", {}) or {}
+        conversation_id = (
+            req.conversation.cid
+            if req is not None and req.conversation is not None
+            else None
+        )
+
+        if agent_runner.was_aborted():
+            status = "aborted"
+        elif final_resp is not None and final_resp.role == "err":
+            status = "error"
+        else:
+            status = "completed"
+
+        await db_helper.insert_provider_stat(
+            umo=event.unified_msg_origin,
+            conversation_id=conversation_id,
+            provider_id=provider_config.get("id", "") or provider.meta().id,
+            provider_model=provider.get_model(),
+            status=status,
+            stats=stats.to_dict(),
+            agent_type="internal",
+        )
+    except Exception as e:
+        logger.warning("Persist provider stats failed: %s", e, exc_info=True)

astrbot/core/pipeline/process_stage/method/agent_sub_stages/third_party.py

@@ -17,7 +17,7 @@ from astrbot.core.agent.runners.deerflow.deerflow_agent_runner import (
 )
 from astrbot.core.agent.runners.dify.dify_agent_runner import DifyAgentRunner
 from astrbot.core.astr_agent_hooks import MAIN_AGENT_HOOKS
-from astrbot.core.message.components import Image
+from astrbot.core.message.components import Image, Record
 from astrbot.core.message.message_event_result import (
     MessageChain,
     MessageEventResult,
@@ -317,8 +317,11 @@ class ThirdPartyAgentSubStage(Stage):
             if isinstance(comp, Image):
                 image_path = await comp.convert_to_base64()
                 req.image_urls.append(image_path)
+            elif isinstance(comp, Record):
+                audio_path = await comp.convert_to_file_path()
+                req.audio_urls.append(audio_path)
 
-        if not req.prompt and not req.image_urls:
+        if not req.prompt and not req.image_urls and not req.audio_urls:
             return
 
         custom_error_message = await self._resolve_persona_custom_error_message(event)
@@ -378,7 +381,7 @@ class ThirdPartyAgentSubStage(Stage):
                 request=req,
                 run_context=AgentContextWrapper(
                     context=astr_agent_ctx,
-                    tool_call_timeout=60,
+                    tool_call_timeout=120,
                 ),
                 agent_hooks=MAIN_AGENT_HOOKS,
                 provider_config=self.prov_cfg,

astrbot/core/pipeline/process_stage/method/star_request.py

@@ -5,6 +5,7 @@ from collections.abc import AsyncGenerator
 from typing import Any
 
 from astrbot.core import logger
+from astrbot.core.i18n import t
 from astrbot.core.message.message_event_result import MessageEventResult
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.star.star import star_map
@@ -62,7 +63,13 @@ class StarRequestSubStage(Stage):
                 )
 
                 if not event.is_stopped() and event.is_at_or_wake_command:
-                    ret = f":(\n\n在调用插件 {md.name} 的处理函数 {handler.handler_name} 时出现异常：{e}"
+                    ret = t(
+                        "pipeline.plugin_handler_error",
+                        locale=self.ctx.get_current_language(),
+                        plugin_name=md.name,
+                        handler_name=handler.handler_name,
+                        error=e,
+                    )
                     event.set_result(MessageEventResult().message(ret))
                     yield
                     event.clear_result()

astrbot/core/pipeline/respond/stage.py

@@ -32,7 +32,6 @@ class RespondStage(Stage):
         Comp.Node: lambda comp: bool(comp.content),  # 转发节点
         Comp.Nodes: lambda comp: bool(comp.nodes),  # 多个转发节点
         Comp.File: lambda comp: bool(comp.file_ or comp.url),
-        Comp.WechatEmoji: lambda comp: comp.md5 is not None,  # 微信表情
         Comp.Json: lambda comp: bool(comp.data),  # Json 卡片
         Comp.Share: lambda comp: bool(comp.url) or bool(comp.title),
         Comp.Music: lambda comp: (

astrbot/core/pipeline/result_decorate/stage.py

@@ -5,7 +5,8 @@ import traceback
 from collections.abc import AsyncGenerator
 
 from astrbot.core import file_token_service, html_renderer, logger
-from astrbot.core.message.components import At, Image, Node, Plain, Record, Reply
+from astrbot.core.i18n import t
+from astrbot.core.message.components import At, Image, Json, Node, Plain, Record, Reply
 from astrbot.core.message.message_event_result import ResultContentType
 from astrbot.core.pipeline.content_safety_check.stage import ContentSafetyCheckStage
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
@@ -275,8 +276,30 @@ class ResultDecorateStage(Stage):
                 and event.get_extra("_llm_reasoning_content")
             ):
                 # inject reasoning content to chain
-                reasoning_content = event.get_extra("_llm_reasoning_content")
-                result.chain.insert(0, Plain(f"🤔 思考: {reasoning_content}\n"))
+                reasoning_content = str(event.get_extra("_llm_reasoning_content"))
+                if event.get_platform_name() == "lark":
+                    result.chain.insert(
+                        0,
+                        Json(
+                            data={
+                                "type": "lark_collapsible_panel_reasoning",
+                                "title": "💭 Thinking",
+                                "expanded": False,
+                                "content": reasoning_content,
+                            },
+                        ),
+                    )
+                else:
+                    result.chain.insert(
+                        0,
+                        Plain(
+                            t(
+                                "pipeline.reasoning_prefix",
+                                locale=self.ctx.get_current_language(),
+                                reasoning_content=reasoning_content,
+                            ),
+                        ),
+                    )
 
             if should_tts and tts_provider:
                 new_chain = []

astrbot/core/pipeline/scheduler.py

@@ -86,10 +86,11 @@ class PipelineScheduler:
         try:
             await self._process_stages(event)
 
-            # 如果没有发送操作, 则发送一个空消息, 以便于后续的处理
+            # 发送一个空消息, 以便于后续的处理
             if isinstance(event, WebChatMessageEvent | WecomAIBotMessageEvent):
                 await event.send(None)
 
             logger.debug("pipeline 执行完毕。")
         finally:
+            event.cleanup_temporary_local_files()
             active_event_registry.unregister(event)

astrbot/core/pipeline/waking_check/stage.py

@@ -1,6 +1,7 @@
 from collections.abc import AsyncGenerator, Callable
 
 from astrbot import logger
+from astrbot.core.i18n import t
 from astrbot.core.message.components import At, AtAll, Reply
 from astrbot.core.message.message_event_result import MessageChain, MessageEventResult
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
@@ -22,6 +23,7 @@ UNIQUE_SESSION_ID_BUILDERS: dict[str, Callable[[AstrMessageEvent], str | None]]
     "qq_official_webhook": lambda e: e.get_sender_id(),
     "lark": lambda e: f"{e.get_sender_id()}%{e.get_group_id()}",
     "misskey": lambda e: f"{e.get_session_id()}_{e.get_sender_id()}",
+    "matrix": lambda e: f"{e.get_sender_id()}_{e.get_group_id() or e.get_session_id()}",
 }
 
 
@@ -186,7 +188,12 @@ class WakingCheckStage(Stage):
                 except Exception as e:
                     await event.send(
                         MessageEventResult().message(
-                            f"插件 {star_map[handler.handler_module_path].name}: {e}",
+                            t(
+                                "pipeline.filter_error",
+                                locale=self.ctx.get_current_language(),
+                                plugin_name=star_map[handler.handler_module_path].name,
+                                error=e,
+                            ),
                         ),
                     )
                     event.stop_event()
@@ -200,7 +207,11 @@ class WakingCheckStage(Stage):
                     if self.no_permission_reply:
                         await event.send(
                             MessageChain().message(
-                                f"您(ID: {event.get_sender_id()})的权限不足以使用此指令。通过 /sid 获取 ID 并请管理员添加。",
+                                t(
+                                    "pipeline.no_permission",
+                                    locale=self.ctx.get_current_language(),
+                                    sender_id=event.get_sender_id(),
+                                ),
                             ),
                         )
                     logger.info(