TradingAgents/tradingagents/agents
Michael Yang 2f870be9a8 feat: route analyst tool loop through Claude Agent SDK (Shape B)
The 4 analysts (market, news, social, fundamentals) now detect a
ChatClaudeAgent LLM and dispatch to an SDK-native runner: LangChain @tool
functions are wrapped as in-process MCP tools via create_sdk_mcp_server, and
the SDK owns the iterative tool-calling loop. Claude returns the final report
in one call, so the analyst node outputs an AIMessage with no tool_calls and
the existing conditional edges route straight to the message-clear step.

Together with the Shape A provider this lets a Claude Max subscription drive
the full TradingAgents graph without an Anthropic API key. Other providers
continue to take the original bind_tools + LangGraph ToolNode path unchanged.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-14 16:08:40 -04:00
..
analysts feat: route analyst tool loop through Claude Agent SDK (Shape B) 2026-04-14 16:08:40 -04:00
managers refactor: remove stale imports, use configurable results path (#499) 2026-04-04 07:35:35 +00:00
researchers refactor: remove stale imports, use configurable results path (#499) 2026-04-04 07:35:35 +00:00
risk_mgmt refactor: remove stale imports, use configurable results path (#499) 2026-04-04 07:35:35 +00:00
trader refactor: remove stale imports, use configurable results path (#499) 2026-04-04 07:35:35 +00:00
utils fix: process all chunk messages for tool call logging, harden memory score normalization (#534, #531) 2026-04-13 07:21:33 +00:00
__init__.py refactor: standardize portfolio manager, five-tier rating scale, fix analyst status tracking 2026-03-22 23:30:29 +00:00