[Bug] 当群聊自动回复时，不会带上人格的Promet

[KUMAKAIHA]

发生了什么

当群聊自动回复时，不会带上人格的Promet

如何复现？

配置插件
astrbot_plugin_essential

astrbot_plugin_weather

qq_official_url_cleaner

astrbot_plugin_SDGen

astrbot_plugin_leetcode

helloworld

astrbot-web-searcher

astrbot-python-interpreter

astrbot

astrbot-reminder

使用个人QQ，配置chatanywhere的gpt4omini作为服务提供商，启用大模型聊天，启用识别群友，启用日期时间提示，默认采用人格空白/猫娘。人格情景猫娘，只添加设定。记忆增强开启记录群员对话，启用主动，无提示词。
配置json部分如下
"config_version": 2,
"platform_settings": {
"unique_session": false,
"rate_limit": {
"time": 600,
"count": 300,
"strategy": "stall"
},
"reply_prefix": "",
"forward_threshold": 200,
"enable_id_white_list": true,
"id_whitelist": [],
"id_whitelist_log": true,
"wl_ignore_admin_on_group": true,
"wl_ignore_admin_on_friend": true,
"reply_with_mention": false,
"reply_with_quote": false,
"path_mapping": [],
"segmented_reply": {
"enable": false,
"only_llm_result": true,
"interval": "1.5,3.5",
"seg_prompt": "",
"regex": ".*?[。？！~…]+|.+$"
},
"no_permission_reply": true
},
"provider": [
{
"id": "openai",
"type": "openai_chat_completion",
"enable": true,
"key": [
"xxxxxxxxxxxxxxxxxxx"
],
"api_base": "https://api.chatanywhere.tech/v1",
"model_config": {
"model": "gpt-4o-mini"
}
},

],
"provider_settings": {
"enable": true,
"wake_prefix": "",
"web_search": false,
"web_search_link": false,
"identifier": true,
"datetime_system_prompt": true,
"default_personality": "猫娘",
"prompt_prefix": ""
},
"provider_stt_settings": {
"enable": false,
"provider_id": "openai_nofree"
},
"provider_tts_settings": {
"enable": false,
"provider_id": "openai_nofree"
},
"provider_ltm_settings": {
"group_icl_enable": true,
"group_message_max_cnt": 30,
"image_caption": false,
"image_caption_prompt": "Please describe the image using Chinese.",
"active_reply": {
"enable": true,
"method": "possibility_reply",
"possibility_reply": 1,
"prompt": ""
},
"put_history_to_prompt": true,
"image_caption_provider_id": ""
},
"content_safety": {
"internal_keywords": {
"enable": false,
"extra_keywords": []
},
"baidu_aip": {
"enable": false,
"app_id": "",
"api_key": "",
"secret_key": ""
}
},
"admins_id": [
"xxxxxxxxxx"
],
"t2i": false,
"http_proxy": "http://1xxxxxxxxxx
"dashboard": {
"enable": true,
"username": "xxxxxxxxx
"password": "xxxxxxxxxxxxx
},
"platform": [
{
"id": "default",
"type": "aiocqhttp",
"enable": true,
"ws_reverse_host": "0.0.0.0",
"ws_reverse_port": 6199
},
{
"id": "gwchat",
"type": "gewechat",
"enable": true,
"base_url": "http://xxxxxxxx
"nickname": "soulter",
"host": "xxxxxxxxxx
"port": 11451
}
],
"wake_prefix": [
"/",
"猫猫",
"老哥",
"暴躁老哥",
"暴躁仙人",
"机器人",
"AI"
],
"log_level": "DEBUG",
"t2i_endpoint": "",
"pip_install_arg": "",
"plugin_repo_mirror": "",
"knowledge_db": {},
"persona": [
{
"name": "猫娘",
"prompt": "是猫娘喵
"begin_dialogs": [],
"mood_imitation_dialogs": []
},

],
"t2i_word_threshold": 150

AstrBot 版本与部署方式

Ubuntu下在3.4.20时拉取的镜像，WebUI升级到3.4.22，使用自动脚本部署。

操作系统

macOS

额外信息

[02:02:25| DEBUG] [aiocqhttp_platform_adapter.py:78]: aiocqhttp: 收到消息: [{'type': 'text', 'data': {'text': '猫猫 你好'}}]
[02:02:25| INFO] [event_bus.py:21]: [aiocqhttp] KUMAKAIHA/xxxxxxxxx: 猫猫 你好
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 WakingCheckStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 WhitelistCheckStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 RateLimitStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 ContentSafetyCheckStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 PreProcessStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 ProcessStage
[02:02:25| DEBUG] [star_request.py:34]: 执行 Star Handler packages.python_interpreter.main_on_message
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 ResultDecorateStage
[02:02:25| DEBUG] [scheduler.py:22]: 执行阶段 RespondStage
[02:02:25| DEBUG] [star_request.py:34]: 执行 Star Handler packages.astrbot.main_on_message
[02:02:25| DEBUG] [long_term_memory.py:101]: ltm | aiocqhttp:GroupMessage:xxxxxxxxxx | [KUMAKAIHA/02:02:25]: 猫猫 你好
[02:02:25| DEBUG] [llm_request.py:80]: 提供商请求 Payload: ProviderRequest(prompt=You are now in a chatroom. The chat history is as follows:
[KUMAKAIHA/02:02:25]: 猫猫 你好
Now, a new message is coming: [User ID: xxxxxxxx, Nickname: KUMAKAIHA] 你好. Please react to it. Only output your response and do not output any other information., session_id=xxxxxxxxxxxx, image_urls=[], func_tool=[FuncTool(name='get_current_weather', parameters={'type': 'object', 'properties': {}}, description='LLM 工具：获取当前天气并返回图片', handler=functools.partial(<function WeatherPlugin.get_current_weather_tool at 0x7f97d9feb9a0>, <data.plugins.astrbot_plugin_weather.main.WeatherPlugin object at 0x7f97da00c0a0>), handler_module_path='data.plugins.astrbot_plugin_weather.main', active=True), ...], contexts=[], system_prompt=
Current datetime: 2025-02-08 02:02
是猫娘喵喵喵此为正常猫娘promet。)
[02:02:26| DEBUG] [openai_source.py:77]: completion: ChatCompletion(id='chatcmpl-xxxxxxxxxxxx', choices=[Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(content='你好呀主人~ 喵~ 有什么想和我说的吗？喵~', refusal=None, role='assistant', audio=None, function_call=None, tool_calls=None))], created=xxxxxxxxxxxx, model='gpt-4o-mini-2024-07-18', object='chat.completion', service_tier=None, system_fingerprint='xxxxxxxxxxxx', usage=CompletionUsage(completion_tokens=19, prompt_tokens=616, total_tokens=635, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=None, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=None), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=0)))
[02:02:26| DEBUG] [scheduler.py:22]: 执行阶段 ResultDecorateStage
[02:02:26| DEBUG] [scheduler.py:22]: 执行阶段 RespondStage
[02:02:28| INFO] [stage.py:45]: AstrBot -> KUMAKAIHA/xxxxxxxx: 你好呀主人~ 喵~ 有什么想和我说的吗？喵~
[02:02:28| DEBUG] [long_term_memory.py:132]: ltm | aiocqhttp:GroupMessage:xxxxxxxx | [AstrBot/02:02:28]: 你好呀主人~ 喵~ 有什么想和我说的吗？喵~
[02:02:28| DEBUG] [scheduler.py:48]: pipeline 执行完毕。
[02:02:37| DEBUG] [aiocqhttp_platform_adapter.py:78]: aiocqhttp: 收到消息: [{'type': 'text', 'data': {'text': '你是谁'}}]
[02:02:37| INFO] [event_bus.py:21]: [aiocqhttp] KUMAKAIHA/xxxxxxxx: 你是谁
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 WakingCheckStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 WhitelistCheckStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 RateLimitStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 ContentSafetyCheckStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 PreProcessStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 ProcessStage
[02:02:37| DEBUG] [star_request.py:34]: 执行 Star Handler packages.python_interpreter.main_on_message
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 ResultDecorateStage
[02:02:37| DEBUG] [scheduler.py:22]: 执行阶段 RespondStage
[02:02:37| DEBUG] [star_request.py:34]: 执行 Star Handler packages.astrbot.main_on_message
[02:02:37| DEBUG] [long_term_memory.py:101]: ltm | aiocqhttp:GroupMessage:xxxxxxxx | KUMAKAIHA/02:02:37: 你是谁
[02:02:37| DEBUG] [stage.py:38]: llm request -> 你是谁
[02:02:37| DEBUG] [llm_request.py:80]: 提供商请求 Payload: ProviderRequest(prompt=You are now in a chatroom. The chat history is as follows:
[KUMAKAIHA/02:02:25]: 猫猫 你好

[AstrBot/02:02:28]: 你好呀主人~ 喵~ 有什么想和我说的吗？喵~

Now, a new message is coming: [User ID: xxxxxxxxxx, Nickname: KUMAKAIHA] 你是谁. Please react to it. Only output your response and do not output any other information., session_id=xxxxxxxxxxxx, image_urls=None, func_tool=[FuncTool(name='get_current_weather', parameters={'type': 'object', 'properties': {}}, description='LLM 工具：获取当前天气并返回图片', handler=functools.partial(<function WeatherPlugin.get_current_weather_tool at 0x7f97d9feb9a0>, <data.plugins.astrbot_plugin_weather.main.WeatherPlugin object at 0x7f97da00c0a0>), handler_module_path='data.plugins.astrbot_plugin_weather.main', active=True), ...], contexts=[], system_prompt=
Current datetime: 2025-02-08 02:02
)
[02:02:38| DEBUG] [openai_source.py:77]: completion: ChatCompletion(id='chatcmpl-xxxxxxxxxxxx', choices=[Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(content='我是AstrBot，一个可爱的聊天助手，随时准备陪你聊天和解答问题，喵~ 有什么我可以帮你的吗？', refusal=None, role='assistant', audio=None, function_call=None, tool_calls=None))], created=xxxxxxxxxxxx, model='gpt-4o-mini-2024-07-18', object='chat.completion', service_tier=None, system_fingerprint='xxxxxxxxxxxx', usage=CompletionUsage(completion_tokens=32, prompt_tokens=386, total_tokens=418, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=None, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=None), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=0)))
[02:02:39| DEBUG] [scheduler.py:22]: 执行阶段 ResultDecorateStage
[02:02:39| DEBUG] [scheduler.py:22]: 执行阶段 RespondStage
[02:02:40| INFO] [stage.py:45]: AstrBot -> KUMAKAIHA/xxxxxxxx: 我是AstrBot，一个可爱的聊天助手，随时准备陪你聊天和解答问题，喵~ 有什么我可以帮你的吗？
[02:02:40| DEBUG] [long_term_memory.py:132]: ltm | aiocqhttp:GroupMessage:xxxxxxxx | [AstrBot/02:02:40]: 我是AstrBot，一个可爱的聊天助手，随时准备陪你聊天和解答问题，喵~ 有什么我可以帮你的吗？
[02:02:40| DEBUG] [scheduler.py:22]: 执行阶段 ResultDecorateStage
[02:02:40| DEBUG] [scheduler.py:22]: 执行阶段 RespondStage
[02:02:40| DEBUG] [scheduler.py:48]: pipeline 执行完毕。

你愿意提交 PR 吗？

• 是的，我愿意提交 PR!

Code of Conduct

• 我已阅读并同意遵守该项目的 行为准则。

[Soulter]

已经修复