Yijia-Xiao 517eeaf4b9 fix(structured): harden structured output for local servers and thinking models ...

- Local servers (LM Studio, vLLM) reject the object-form tool_choice langchain
  sends for function calling. The generic openai_compatible provider now binds
  the schema as a tool without forcing tool_choice.
- A structured call can return no parsed result (a thinking model answering in
  plain text); fall back to free text with a clear reason instead of an opaque
  render error.

2026-06-21 22:09:43 +00:00

..

agent_states.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

agent_utils.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

core_stock_tools.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

fundamental_data_tools.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

macro_data_tools.py

feat(data): add FRED macro indicators as an optional vendor

2026-06-14 06:08:31 +00:00

market_data_validation_tools.py

feat(market): verified market-data snapshot to ground numeric claims

2026-05-31 01:58:32 +00:00

memory.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

news_data_tools.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

prediction_markets_tools.py

feat(data): add Polymarket prediction markets as a keyless vendor

2026-06-14 06:30:43 +00:00

rating.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00

structured.py

fix(structured): harden structured output for local servers and thinking models

2026-06-21 22:09:43 +00:00

technical_indicators_tools.py

chore(lint): make the repository ruff-clean under the strict select

2026-06-14 16:38:36 +00:00