lorenzejay e48cf3604e refactor: implement _loop_response_model to manage response_model during tool loops ...

This change introduces the _loop_response_model method in BaseAgentExecutor to conditionally return the response_model based on the presence of tools. The CrewAgentExecutor and AgentExecutor classes have been updated to utilize this new method, ensuring that the response_model is not sent to the LLM during tool loops, which prevents issues with structured-output schemas. Additionally, tests have been added to verify the correct behavior of this implementation.

2026-04-24 12:20:32 -07:00

..

agent_adapters

chore: extension points

2026-02-05 12:49:54 -05:00

agent_builder

feat: async task support (#4024)

2025-12-04 13:34:29 -08:00

__init__.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_a2a_trust_completion_status.py

feat: add additional a2a events and enrich event metadata

2026-01-16 16:57:31 -05:00

test_agent_a2a_kickoff.py

feat: add a2a liteagent, auth, transport negotiation, and file support

2026-01-30 17:10:00 -05:00

test_agent_a2a_wrapping.py

Gl/feat/a2a refactor (#3793)

2025-10-31 18:42:03 -07:00

test_agent_executor.py

refactor: implement _loop_response_model to manage response_model during tool loops

2026-04-24 12:20:32 -07:00

test_agent_inject_date.py

refactor: deduplicate sync/async task execution and kickoff in agent

2026-03-25 19:39:42 +08:00

test_agent_reasoning.py

Lorenze/feat/plan execute pattern (#4817)

2026-03-15 18:33:17 -07:00

test_agent.py

refactor: use shared I18N_DEFAULT singleton

2026-04-09 04:29:53 +08:00

test_async_agent_executor.py

feat: runtime state checkpointing, event system, and executor refactor

2026-04-07 03:22:30 +08:00

test_crew_agent_parser.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_lite_agent.py

chore: add deprecation decorator to LiteAgent

2026-04-14 00:51:11 +08:00

test_native_tool_calling.py

feat: runtime state checkpointing, event system, and executor refactor

2026-04-07 03:22:30 +08:00