Devin AI d93040efa7 fix(#5878): preserve reasoning_content from DeepSeek thinking mode ...

Extract reasoning_content from litellm response and store it on the
LLM instance so that executors can propagate it into conversation
history as required by the DeepSeek API.

Changes:
- LLM._handle_non_streaming_response: extract reasoning_content from
  the response message and store it as self.reasoning_content
- LLM.call: reset reasoning_content at the start of each call
- format_message_for_llm: accept optional reasoning_content param;
  include it in assistant messages only
- LLMMessage TypedDict: add reasoning_content field
- CrewAgentExecutor: pass reasoning_content through _append_message
  for both sync and async loops (ReAct + native tools)
- AgentExecutor (experimental): same propagation in
  _append_message_to_state for native tools path

Tests: 13 new tests covering LLM extraction, format_message_for_llm,
and executor integration.

Co-Authored-By: João <joao@crewai.com>

2026-05-21 05:49:01 +00:00

..

cli

feat: Skills Repository — registry, cache, CLI, and SDK integration (#5867)

2026-05-20 14:38:25 -03:00

crewai

fix(#5878): preserve reasoning_content from DeepSeek thinking mode

2026-05-21 05:49:01 +00:00

crewai-core

feat: Skills Repository — registry, cache, CLI, and SDK integration (#5867)

2026-05-20 14:38:25 -03:00

crewai-files

feat: bump versions to 1.14.5

2026-05-19 03:11:26 +08:00

crewai-tools

feat: bump versions to 1.14.5

2026-05-19 03:11:26 +08:00

devtools

feat: generate categorized release notes for enterprise

2026-05-20 00:24:26 +08:00