- MCPTool_T must be imported at runtime, not in TYPE_CHECKING
- Add __future__ annotations for forward references
- Fix remaining conflict markers in openai_source.py
Resolved conflicts:
- openai_source.py: keep dev version with abort_signal filtering
- customizer.ts: keep dev version with viewMode functionality
- useSessions.ts: keep dev version with pendingSessionId handling
- platformUtils.js: keep dev version with correct tutorial links
- AddNewPlatform.vue: keep dev version with correct docs link
- FullLayout.vue: keep dev version with viewMode-based logic
- VerticalHeader.vue: keep dev version with viewMode-based logic
* feat: add two-phase startup lifecycle
Allow the dashboard to become available before plugin bootstrap completes and surface runtime readiness and failure states to API callers.
Guard plugin-facing endpoints until runtime is ready and clean up provider and plugin runtime state safely across bootstrap failures, retries, stop, and restart flows.
* fix: harden runtime cleanup review fixes
Continue terminating remaining providers and disable MCP servers even if one provider terminate hook fails.
Also add InitialLoader failure-path coverage and extract guarded plugin routes into a shared constant for easier review and maintenance.
* fix: harden deferred startup recovery
* fix: streamline runtime guard handling
* fix: simplify runtime lifecycle coordination
* fix: restore orchestrator logger binding
The internal ToolSet (base.py) was missing add_tool() and merge()
methods that the agent code expects. When tmgr.get_full_tool_set()
returned a base.py ToolSet, calls to add_tool() and merge() failed.
Added:
- add_tool() as alias to add()
- merge() method to merge another ToolSet
This fixes runtime crash: AttributeError: 'ToolSet' object has no attribute 'add_tool'
- Add _message_count and _last_activity_timestamp to orchestrator
- Add record_activity() method to orchestrator
- Add name field to get_protocol_status returns
- Add total_messages and last_activity to get_stats
- Update tests to verify new fields
- Fix orchestrator to use anyio.get_cancelled_exc_class() instead of anyio.CancelledError
- Fix tests to properly check for anyio compliance (not violations)
- Add type annotations for MCP exception fallbacks in registry.py
- Remove unused type: ignore comment in mcp/tool.py
- All 111 tests pass
- uvx ty check passes
- ruff check passes
Implement the new _internal package structure for AstrBot runtime:
- Add AstrbotOrchestrator with LSP, MCP, ACP, ABP protocol clients
- Add AstrbotGateway server with WebSocket support
- Add comprehensive test suite for runtime module
- Add tools base module for MCP tools
Implements bootstrap function using anyio task groups for
concurrent protocol client initialization.
- Add version check at startup in both __main__.py and cmd_run.py
- Suggest using `uv run -m astrbot` or reinstalling with uv
- Add ABC base class and abstract methods to ComputerBooter
- Improve type annotations in OpenAIAgentsRunner
Move the openai-agents SDK integration from core/agent/runners/openai_agents
to _internal/agents/openai_agents to follow the internal implementation pattern.
This commit adds the _internal package structure for AstrBot's
standardized MCP & Skills support:
astrbot/_internal/mcp/:
- MCPClient for MCP server connections
- MCPTool wrapper for MCP tools
- MCP configuration management
astrbot/_internal/skills/:
- SkillManager for skill lifecycle
- Skill parser and loader
- SkillToToolConverter for tool-based skills
- Prompt builder for skills
astrbot/_internal/tools/:
- ToolSchema, FunctionTool, ToolSet base definitions
- FunctionToolManager for tool registry
- Builtin tools (cron, send_message, kb_query)
- Tool providers (internal, plugin, computer)
astrbot/api/:
- Public API for tools (ToolRegistry, tool decorator)
- Public API for MCP (get_mcp_servers, register_mcp_server)
- Public API for skills (get_skill_manager, skill_to_tool)
