Yufeng He 2cb6c84eeb feat: context token counting support for multimodal content (images, audio, and chain-of-thought) (#6596) ...

EstimateTokenCounter 之前只计算 TextPart，完全忽略 ImageURLPart、
AudioURLPart 和 ThinkPart。多模态对话中图片占 500-2000 token，
不被计入会导致 context 压缩触发过晚，API 先报 context_length_exceeded。

改动：
- ImageURLPart 按 765 token 估算（OpenAI vision 低/高分辨率中位数）
- AudioURLPart 按 500 token 估算
- ThinkPart 的文本内容正常计算
- 10 个新测试覆盖各类型单独和混合场景

Co-authored-by: Yufeng He <40085740+universeplayer@users.noreply.github.com>

2026-03-19 17:25:49 +08:00

..

agent

feat: context token counting support for multimodal content (images, audio, and chain-of-thought) (#6596)

2026-03-19 17:25:49 +08:00

fixtures

feat(telegram): supports sendMessageDraft API (#5726)

2026-03-05 11:20:28 +08:00

test_kook

perf: Implement Pydantic data models for the KOOK adapter to enhance data retrieval and message schema validation (#5719)

2026-03-17 18:05:58 +08:00

unit

fix: subagent lookup failure when using default persona (#5672)

2026-03-17 15:42:15 +08:00

conftest.py

test: enhance test framework with comprehensive fixtures and mocks (#5354)

2026-02-23 23:35:15 +08:00

test_api_key_open_api.py

test: add tests for star base class and config management (#5356)

2026-03-01 00:06:04 +08:00

test_backup.py

fix: harden backup import for duplicate platform stats (#5594)

2026-03-01 20:46:35 +09:00

test_chat_route.py

fix: avoid webchat stream result crash on queue errors (#6123)

2026-03-14 23:41:28 +08:00

test_computer_config.py

test(computer): add tests for credentials discovery and config logging

2026-02-19 01:26:04 +08:00

test_computer_skill_sync.py

feat(skills): merge sandbox built-ins with uploaded skill sync

2026-02-13 03:20:51 +08:00

test_dashboard.py

fix: subagent lookup failure when using default persona (#5672)

2026-03-17 15:42:15 +08:00

test_kb_import.py

feat: add pre-chunk import strategy for knowledge base (#3973)

2025-12-12 23:15:11 +08:00

test_local_shell_component.py

fix: improve Windows local skill file reading (#6028)

2026-03-11 23:58:28 +09:00

test_main.py

Merge remote-tracking branch 'origin/master' into feat/neo-skill-self-iteration

2026-03-02 09:37:06 +08:00

test_neo_skill_sync.py

feat(skills): add neo lifecycle tools and stable sync manager

2026-02-11 17:14:47 +08:00

test_neo_skill_tools.py

feat(skills): add neo lifecycle tools and stable sync manager

2026-02-11 17:14:47 +08:00

test_openai_source.py

fix: drop Groq reasoning_content from assistant history (#6065)

2026-03-15 22:51:52 +08:00

test_pip_helper_modules.py

fix: install only missing plugin dependencies (#6088)

2026-03-12 11:50:29 +09:00

test_pip_installer.py

fix: set packaged Windows runtime build env for pip native builds (#6575)

2026-03-19 00:07:03 +09:00

test_plugin_manager.py

fix: install only missing plugin dependencies (#6088)

2026-03-12 11:50:29 +09:00

test_profile_aware_tools.py

test(computer): add profile-aware sandbox selection tests

2026-02-21 01:03:58 +08:00

test_quoted_message_parser.py

fix: 完善转发引用解析与图片回退并支持配置化控制 (#5054)

2026-02-12 23:42:29 +09:00

test_runtime_env.py

fix: detect desktop runtime without frozen python (#5859)

2026-03-07 21:42:56 +09:00

test_security_fixes.py

Replace insecure random with secrets module in cryptographic contexts (#3248)

2025-11-03 02:43:00 +08:00

test_skill_manager_sandbox_cache.py

test(skill_manager): update sandbox cache path expectations (#5706)

2026-03-05 22:47:20 +08:00

test_skill_metadata_enrichment.py

fix: parse multiline frontmatter description in SKILL.md (#6460)

2026-03-17 13:53:16 +08:00

test_smoke.py

test: add tests for star base class and config management (#5356)

2026-03-01 00:06:04 +08:00

test_temp_dir_cleaner.py

feat: temporary file handling and introduce TempDirCleaner (#5026)

2026-02-12 01:04:48 +08:00

test_tool_loop_agent_runner.py

test: add tests for star base class and config management (#5356)

2026-03-01 00:06:04 +08:00