[Bug]使用 DeepSeek Reasoner 开启工具调用时报 400 错误

[lingyun14beta]

What happened / 发生了什么

使用 DeepSeek Reasoner 模型开启工具调用时，报 400 错误导致对话失败：
Error code: 400 - {'error': {'message': 'Invalid assistant message: content or tool_calls must be set'}}
原因是 openai_source.py 的 _query_stream 方法缺少空 assistant 消息过滤逻辑。DeepSeek Reasoner 工具调用时只返回 reasoning_content 无 completion_text，序列化后 content 为空，触发 DeepSeek 严格校验。_query 在 #7202 已有修复，但 _query_stream 未同步，且 _query 也未覆盖 content == [] 的情况。

Reproduce / 如何复现？

1.Provider 设置为 DeepSeek，模型选择 deepseek-reasoner
2.开启工具调用功能
3.发送消息触发工具调用
4.调用完成后再次发送任意对话信息
5.400报错

AstrBot version, deployment method (e.g., Windows Docker Desktop deployment), provider used, and messaging platform used. / AstrBot 版本、部署方式（如 Windows Docker Desktop 部署）、使用的提供商、使用的消息平台适配器

win10 AstrBotv4.23.2

OS

Windows

Logs / 报错日志

[2026-04-22 13:49:47.185] [Core] [INFO] [core.event_bus:61]: [default] [default(aiocqhttp)] 凌/qq号: 你好
[2026-04-22 13:49:47.223] [Core] [INFO] [astrbot_plugin_qq_group_sign.main:376]: 成功捕获 aiocqhttp 机器人实例，后台 API 调用已启用。
[2026-04-22 13:49:47.589] [Core] [WARN] [v4.23.2] [runners.tool_loop_agent_runner:510]: Chat Model deepseek/deepseek-reasoner request error: Error code: 400 - {'error': {'message': 'Invalid assistant message: content or tool_calls must be set', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
Traceback (most recent call last):
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\agent\runners\tool_loop_agent_runner.py", line 465, in iter_llm_responses_with_fallback
async for attempt in retrying:
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\tenacity\asyncio_init.py", line 166, in anext
do = await self.iter(retry_state=self.retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\tenacity\asyncio_init.py", line 153, in iter
result = await action(retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\tenacity_utils.py", line 99, in inner
return call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\tenacity_init_.py", line 400, in
self._add_action_func(lambda rs: rs.outcome.result())
^^^^^^^^^^^^^^^^^^^
File "C:\Program Files\Python311\Lib\concurrent\futures_base.py", line 449, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "C:\Program Files\Python311\Lib\concurrent\futures_base.py", line 401, in __get_result
raise self._exception
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\agent\runners\tool_loop_agent_runner.py", line 469, in _iter_llm_responses_with_fallback
async for resp in self._iter_llm_responses(
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\agent\runners\tool_loop_agent_runner.py", line 429, in _iter_llm_responses
async for resp in stream: # type: ignore
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\provider\sources\openai_source.py", line 1235, in text_chat_stream
) = await self._handle_api_error(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\provider\sources\openai_source.py", line 1109, in _handle_api_error
raise e
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\provider\sources\openai_source.py", line 1222, in text_chat_stream
async for response in self._query_stream(payloads, func_tool):
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\astrbot\core\provider\sources\openai_source.py", line 622, in _query_stream
stream = await self.client.chat.completions.create(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\openai\resources\chat\completions\completions.py", line 2678, in create
return await self._post(
^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\openai_base_client.py", line 1881, in post
return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator\Desktop\chatbot\AstrBotLauncher-0.2.0\AstrBot\venv\Lib\site-packages\openai_base_client.py", line 1666, in request
raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': 'Invalid assistant message: content or tool_calls must be set', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
[2026-04-22 13:49:47.755] [Core] [INFO] [respond.stage:183]: Prepare to send - 凌/qq号： [ComponentType.Node]

Are you willing to submit a PR? / 你愿意提交 PR 吗？

• Yes!

Code of Conduct

• I have read and agree to abide by the project's Code of Conduct。

[dosubot]

Hi @lingyun14beta! I'm Dosu and I’m helping the AstrBot team.

确认了这个 Bug。问题分析如下：

根本原因： PR #7202 在 _query（非流式）方法中添加了空 assistant 消息过滤逻辑 [1]，但 _query_stream（流式）方法没有同步这个修复 [2]。DeepSeek Reasoner 工具调用后只返回 reasoning_content，content 为空，流式请求时这条空消息直接发给了 API，触发 400 错误。

_query 中现有的过滤逻辑 [3]：

if not tool_calls and (content == "" or content is None):
    # 过滤空 assistant 消息

需要修复的两个点：

1. _query_stream 需要在调用 API 前加入与 _query 相同的过滤逻辑 [4]
2. _query 和 _query_stream 都应补充 content == [] 的判断，覆盖 content 为空列表的情况 [5]

建议将过滤逻辑提取为一个共享方法，避免两处重复代码。期待你的 PR！

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.