Fixed IndentationError in trading_graph.py within the _get_provider_kwargs method.
Optimized logic to fetch the configuration value once instead of twice.
Removed temporary French comments.
Updated anthropic_client.py to include thinking_config in the allowed_keys whitelist so the parameters are properly passed to the LLM.
Update _get_provider_kwargs() in trading_graph.py to support extended reasoning for Claude models.
Dynamically injects the thinking_config dictionary when the provider is 'anthropic' and a budget is set.
Keeps the implementation consistent with existing OpenAI and Google logic.
Addresses the missing parameter issue mentioned in llm_clients/TODO.md
InvestDebateState was missing bull_history, bear_history, judge_decision.
RiskDebateState was missing aggressive_history, conservative_history,
neutral_history, latest_speaker, judge_decision. This caused KeyError
in _log_state() and reflection, especially with edge-case config values.
Prevents UnicodeEncodeError on Windows where the default encoding
(cp1252/gbk) cannot handle Unicode characters in LLM output.
Closes#77, closes#114, closes#126, closes#215, closes#332
- Add StatsCallbackHandler for tracking LLM calls, tool calls, and tokens
- Integrate callbacks into TradingAgentsGraph and all LLM clients
- Dynamic agent/report counts based on selected analysts
- Fix report completion counting (tied to agent completion)
Adel Janbeih
Yijia-Xiao
makk9
luohy15
mirza-samad-ahmed-baig
Edward Sun
maxer137