hetaoBackend fa9fba3f8e fix: preserve reasoning_content in multi-turn conversations ...

When using thinking-enabled models (like Kimi K2.5, DeepSeek), the API
expects reasoning_content on all assistant messages. The original
ChatDeepSeek stores reasoning_content in additional_kwargs but doesn't
include it when making subsequent API calls, causing "reasoning_content
is missing" errors.

This adds PatchedChatDeepSeek which overrides _get_request_payload to
restore reasoning_content from additional_kwargs into the payload.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

2026-01-28 14:04:35 +08:00

..

agents

fix: preserve reasoning_content in multi-turn conversations

2026-01-28 14:04:35 +08:00

community

feat: fallback to error reporting

2026-01-28 12:23:29 +08:00

config

feat: add environment variable injection for Docker sandbox

2026-01-24 22:36:05 +08:00

gateway

fix: ensure MCP and skills config changes are immediately reflected

2026-01-25 22:40:09 +08:00

mcp

fix: ensure MCP and skills config changes are immediately reflected

2026-01-25 22:40:09 +08:00

models

fix: preserve reasoning_content in multi-turn conversations

2026-01-28 14:04:35 +08:00

reflection

chore: use ruff to lint and auto-format

2026-01-14 09:08:20 +08:00

sandbox

feat: add podcast generation skill

2026-01-26 13:16:35 +08:00

skills

fix: ensure MCP and skills config changes are immediately reflected

2026-01-25 22:40:09 +08:00

tools

fix: ensure MCP and skills config changes are immediately reflected

2026-01-25 22:40:09 +08:00

utils

feat: add thread-safety and graceful shutdown to AioSandboxProvider (#7)

2026-01-16 22:28:19 +08:00

__init__.py

chore: add an empty __init__.py

2026-01-14 07:16:27 +08:00