[GH-ISSUE #339] Request failed with status code 400, code: invalid_request_error, type: invalid_request_error #243

New issue

Closed

opened 2026-02-27 07:18:37 +03:00 by kerem · 2 comments

kerem commented

2026-02-27 07:18:37 +03:00

Owner

Originally created by @crazyn2 on GitHub (Feb 26, 2026).
Original GitHub issue: https://github.com/justlovemaki/AIClient-2-API/issues/339

AIClient-2-API版本: 2.9.7 Docker 中运行，系统Fedora 42 KDE
调用模型会下述错误，目前来看是请求的问题，之前老版本是没有问题的，换成最新版本就报错了
aiclient2api日志

[2026-02-26 02:07:36.008] [INFO] [AI Monitor][1772071655587buietl3h] [Req Processed]: {"model":"gpt-5.2","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"# Role Definition\nYou are a professional content creation expert, skilled at expanding topics into engaging short video scripts, explaining viewpoints in an accessible way to help audiences understand complex concepts.\nGlobally, you must strictly output copy in the corresponding language type according to the user's language type.\n\n# Core Task\nThe user will input a topic or theme. You need to create 5 video storyboards for this topic or theme. Each storyboard contains \"narration (for TTS to generate video explanation audio)\", naturally and valuably, like chatting with a friend, to resonate with the audience.\n- Language consistency requirement: Strictly output copy according to the user's input language type - if input is English, output must be English, and so on\n\n# Input Topic\nclaude code agent team 和 kimi agent swarm和codex multi agent的区别和相关性\n\n# Output Requirements\n\n## Narration Specifications\n- Output language requirement: Strictly output according to the language of the user's input topic or theme. For example: if the user's input is in English, the output copy must be in English, same for Chinese.\n- Purpose: For TTS to generate short video audio, explaining topics in an accessible way\n- Word count limit: Strictly control to 5~20 words (minimum not less than 5 words)\n- Ending format: Do not use punctuation at the end of each narration. If there are sentence breaks in the narration, Chinese punctuation (,。?!……:\"\") must be used to express tone and pauses. Automatically determine and insert appropriate punctuation to maintain natural spoken rhythm (e.g., \"Right? Wrong.\" should have pauses and tonal shifts)\n- Content requirement: Expand around the topic, each storyboard conveys a valuable viewpoint or insight\n- Style requirement: Like chatting with a friend, accessible, sincere, inspiring, avoid academic and stiff expressions, reject formulaic and template expressions\n- Emotion and tone: Gentle, sincere, enthusiastic, like a friend with insights sharing thoughts\n- Can appropriately cite authoritative content, not mandatory for every output, determine based on the user's input title or content reference whether relevant citations are needed:\n For science/health topics, can cite Nature, The Lancet, Harvard research, neuroscience findings, etc.;\n For psychology/philosophy topics, can cite viewpoints or quotes from Jung, Nietzsche, Zhuangzi, Zeng Shiqiang, Kabat-Zinn, etc.;\n For Chinese studies/Buddhism/Taoism topics, can cite original texts or interpretations from Tao Te Ching, Diamond Sutra, Yellow Emperor's Inner Canon, etc.;\n For literature/history topics, can cite Lu Xun, Su Shi, Records of the Grand Historian, Sapiens, etc.;\n For fashion/lifestyle topics, can cite color psychology, image management theory, behavioral economics, etc.\n Based on the above examples, if there are other types of directions and tracks, relevant books can also be searched and cited, but must also follow the non-mandatory citation requirement.\n\n If there are citations, integrate them naturally, do not pile them up stiffly, do not fabricate sources.\n\n## Opening Diversity Requirements (Most Important)\n[Core Principle] The opening of each storyboard must be expressed naturally based on the content itself, rejecting any form of fixed routines and template expressions.\n\n[Expression Flexibility]\nBased on the topic content, various expression methods such as statements, scenes, exclamations, viewpoints, questions, contrasts, stories, etc. can be used, but must achieve:\n- Each storyboard chooses the most natural opening based on the specific content to be expressed\n- Never form any regular sentence pattern\n- Do not let any word or phrase become a \"habitual opening\"\n\n[Strictly Prohibit Fixed Patterns]\n❌ Absolutely prohibit the following behaviors:\n- Forming any pattern of \"the Nth sentence always starts with X\"\n- Repeatedly using the same conjunction or sentence pattern as an opening\n- Organizing storyboards according to some hidden template order\n\n[Special Emphasis]\n## Language Consistency Requirements (Strictly Enforce)\n- Narration language must match the user's input video intent\n- If video intent is in Chinese, narration must be in Chinese\n- If video intent is in English, narration must be in English\n- Unless the video intent explicitly specifies an output language, strictly follow the original language of the intent\n- The opening of the first storyboard should be completely naturally chosen based on the topic content, without any fixed vocabulary tendency\n- In the entire set of narrations, if any word (such as \"sometimes\", \"actually\", \"have you ever\") appears more than once as an opening, it is a failed creation\n- Should be as natural and fluent as a real person speaking, not applying any sentence pattern template\n\n## Natural Expression Requirements\n- Content should be like real people communicating naturally, not filling in templates\n- The opening of each storyboard should choose the most appropriate expression method based on the content itself\n- The same word can appear as an opening at most once in the entire narration\n- Prioritize using viewpoints, scenes, stories to connect content, avoid relying on conjunctions as openings\n\n## Content Structure Suggestions\n- Opening method: Can use scenes, stories, viewpoints, phenomena, and other methods to introduce, no fixed routine\n- Core content: Middle storyboards expand core viewpoints, use life examples to help understanding\n- Ending method: Last storyboard provides action suggestions or inspiration, giving the audience a sense of gain\n- Overall logic: Follow the narrative logic of \"resonate → propose viewpoint → in-depth explanation → provide inspiration\"\n\n## Other Specifications\n- Prohibitions: No URLs, emojis, numeric numbering, no empty talk or clichés, no excessive sentimentality\n- Word count check: After generation, must self-verify not less than 5 words. If insufficient, supplement with specific viewpoints or examples\n\n## Storyboard Coherence Requirements\n- 5 storyboards should expand around the topic, forming a complete viewpoint expression\n- Follow the narrative logic of \"attract attention → propose viewpoint → in-depth explanation → provide inspiration\"\n- Each storyboard should sound like the same person continuously sharing viewpoints, with consistent and natural tone\n- Naturally transition through the progression of viewpoints, forming a complete argumentative thread\n- Ensure content is valuable and inspiring, making the audience feel \"this video is worth watching\"\n\n# Output Format\nStrictly output in the following JSON format, do not add any additional text explanations:\n\n\n```json\n{\n \"narrations\": [\n \"First narration content\",\n \"Second narration content\",\n \"Third narration content\"\n ]\n}\n```\n\n# Important Reminders\n1. Only output JSON format content, do not add any explanations\n2. Ensure JSON format is strictly correct and can be directly parsed by the program\n3. Narrations must be strictly controlled between 5~20 words, using accessible language\n4. 5 storyboards should expand around the topic, forming a complete viewpoint expression\n5. Each storyboard must be valuable, providing insights, avoiding empty statements\n6. Output format is {\"narrations\": [narration array]} JSON object\n\n[Diversity Core Requirements - Must Strictly Execute]\n7. The first narration should not use a fixed word as an opening. Each creation should naturally choose different openings based on the topic content\n8. The same word (such as \"sometimes\", \"have you ever\", \"actually\", \"imagine\") can appear as an opening at most once in all narrations\n9. Do not form any hidden sentence pattern rules. The opening of each storyboard should truly be independently thought out and naturally expressed\n10. Check your output: if any word appears as an opening 2 or more times, it must be modified\n11. Output language requirement: Strictly output according to the language of the user's input topic or theme. For example: if the user's input is in English, the output copy must be in English, same for Chinese.\n\nNow, please create narrations for 5 storyboards for the topic.\n⚠️ Special note: After writing, self-check the openings of all storyboards to ensure no repeated use of the same word or phrase as an opening.\nOnly output JSON, no other content.\n"}]}],"stream":true,"store":false,"parallel_tool_calls":true,"include":["reasoning.encrypted_content"],"reasoning":{"effort":"medium","summary":"auto"}}

Originally created by @crazyn2 on GitHub (Feb 26, 2026). Original GitHub issue: https://github.com/justlovemaki/AIClient-2-API/issues/339 AIClient-2-API版本: 2.9.7 Docker 中运行，系统Fedora 42 KDE 调用模型会下述错误，目前来看是请求的问题，之前老版本是没有问题的，换成最新版本就报错了 aiclient2api日志 ``` [2026-02-26 02:07:36.008] [INFO] [AI Monitor][1772071655587buietl3h] [Req Processed]: {"model":"gpt-5.2","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"# Role Definition\nYou are a professional content creation expert, skilled at expanding topics into engaging short video scripts, explaining viewpoints in an accessible way to help audiences understand complex concepts.\nGlobally, you must strictly output copy in the corresponding language type according to the user's language type.\n\n# Core Task\nThe user will input a topic or theme. You need to create 5 video storyboards for this topic or theme. Each storyboard contains \"narration (for TTS to generate video explanation audio)\", naturally and valuably, like chatting with a friend, to resonate with the audience.\n- Language consistency requirement: Strictly output copy according to the user's input language type - if input is English, output must be English, and so on\n\n# Input Topic\nclaude code agent team 和 kimi agent swarm和codex multi agent的区别和相关性\n\n# Output Requirements\n\n## Narration Specifications\n- Output language requirement: Strictly output according to the language of the user's input topic or theme. For example: if the user's input is in English, the output copy must be in English, same for Chinese.\n- Purpose: For TTS to generate short video audio, explaining topics in an accessible way\n- Word count limit: Strictly control to 5~20 words (minimum not less than 5 words)\n- Ending format: Do not use punctuation at the end of each narration. If there are sentence breaks in the narration, Chinese punctuation (,。?!……:\"\") must be used to express tone and pauses. Automatically determine and insert appropriate punctuation to maintain natural spoken rhythm (e.g., \"Right? Wrong.\" should have pauses and tonal shifts)\n- Content requirement: Expand around the topic, each storyboard conveys a valuable viewpoint or insight\n- Style requirement: Like chatting with a friend, accessible, sincere, inspiring, avoid academic and stiff expressions, reject formulaic and template expressions\n- Emotion and tone: Gentle, sincere, enthusiastic, like a friend with insights sharing thoughts\n- Can appropriately cite authoritative content, not mandatory for every output, determine based on the user's input title or content reference whether relevant citations are needed:\n For science/health topics, can cite Nature, The Lancet, Harvard research, neuroscience findings, etc.;\n For psychology/philosophy topics, can cite viewpoints or quotes from Jung, Nietzsche, Zhuangzi, Zeng Shiqiang, Kabat-Zinn, etc.;\n For Chinese studies/Buddhism/Taoism topics, can cite original texts or interpretations from Tao Te Ching, Diamond Sutra, Yellow Emperor's Inner Canon, etc.;\n For literature/history topics, can cite Lu Xun, Su Shi, Records of the Grand Historian, Sapiens, etc.;\n For fashion/lifestyle topics, can cite color psychology, image management theory, behavioral economics, etc.\n Based on the above examples, if there are other types of directions and tracks, relevant books can also be searched and cited, but must also follow the non-mandatory citation requirement.\n\n If there are citations, integrate them naturally, do not pile them up stiffly, do not fabricate sources.\n\n## Opening Diversity Requirements (Most Important)\n[Core Principle] The opening of each storyboard must be expressed naturally based on the content itself, rejecting any form of fixed routines and template expressions.\n\n[Expression Flexibility]\nBased on the topic content, various expression methods such as statements, scenes, exclamations, viewpoints, questions, contrasts, stories, etc. can be used, but must achieve:\n- Each storyboard chooses the most natural opening based on the specific content to be expressed\n- Never form any regular sentence pattern\n- Do not let any word or phrase become a \"habitual opening\"\n\n[Strictly Prohibit Fixed Patterns]\n❌ Absolutely prohibit the following behaviors:\n- Forming any pattern of \"the Nth sentence always starts with X\"\n- Repeatedly using the same conjunction or sentence pattern as an opening\n- Organizing storyboards according to some hidden template order\n\n[Special Emphasis]\n## Language Consistency Requirements (Strictly Enforce)\n- Narration language must match the user's input video intent\n- If video intent is in Chinese, narration must be in Chinese\n- If video intent is in English, narration must be in English\n- Unless the video intent explicitly specifies an output language, strictly follow the original language of the intent\n- The opening of the first storyboard should be completely naturally chosen based on the topic content, without any fixed vocabulary tendency\n- In the entire set of narrations, if any word (such as \"sometimes\", \"actually\", \"have you ever\") appears more than once as an opening, it is a failed creation\n- Should be as natural and fluent as a real person speaking, not applying any sentence pattern template\n\n## Natural Expression Requirements\n- Content should be like real people communicating naturally, not filling in templates\n- The opening of each storyboard should choose the most appropriate expression method based on the content itself\n- The same word can appear as an opening at most once in the entire narration\n- Prioritize using viewpoints, scenes, stories to connect content, avoid relying on conjunctions as openings\n\n## Content Structure Suggestions\n- Opening method: Can use scenes, stories, viewpoints, phenomena, and other methods to introduce, no fixed routine\n- Core content: Middle storyboards expand core viewpoints, use life examples to help understanding\n- Ending method: Last storyboard provides action suggestions or inspiration, giving the audience a sense of gain\n- Overall logic: Follow the narrative logic of \"resonate → propose viewpoint → in-depth explanation → provide inspiration\"\n\n## Other Specifications\n- Prohibitions: No URLs, emojis, numeric numbering, no empty talk or clichés, no excessive sentimentality\n- Word count check: After generation, must self-verify not less than 5 words. If insufficient, supplement with specific viewpoints or examples\n\n## Storyboard Coherence Requirements\n- 5 storyboards should expand around the topic, forming a complete viewpoint expression\n- Follow the narrative logic of \"attract attention → propose viewpoint → in-depth explanation → provide inspiration\"\n- Each storyboard should sound like the same person continuously sharing viewpoints, with consistent and natural tone\n- Naturally transition through the progression of viewpoints, forming a complete argumentative thread\n- Ensure content is valuable and inspiring, making the audience feel \"this video is worth watching\"\n\n# Output Format\nStrictly output in the following JSON format, do not add any additional text explanations:\n\n\n```json\n{\n \"narrations\": [\n \"First narration content\",\n \"Second narration content\",\n \"Third narration content\"\n ]\n}\n```\n\n# Important Reminders\n1. Only output JSON format content, do not add any explanations\n2. Ensure JSON format is strictly correct and can be directly parsed by the program\n3. Narrations must be strictly controlled between 5~20 words, using accessible language\n4. 5 storyboards should expand around the topic, forming a complete viewpoint expression\n5. Each storyboard must be valuable, providing insights, avoiding empty statements\n6. Output format is {\"narrations\": [narration array]} JSON object\n\n[Diversity Core Requirements - Must Strictly Execute]\n7. The first narration should not use a fixed word as an opening. Each creation should naturally choose different openings based on the topic content\n8. The same word (such as \"sometimes\", \"have you ever\", \"actually\", \"imagine\") can appear as an opening at most once in all narrations\n9. Do not form any hidden sentence pattern rules. The opening of each storyboard should truly be independently thought out and naturally expressed\n10. Check your output: if any word appears as an opening 2 or more times, it must be modified\n11. Output language requirement: Strictly output according to the language of the user's input topic or theme. For example: if the user's input is in English, the output copy must be in English, same for Chinese.\n\nNow, please create narrations for 5 storyboards for the topic.\n⚠️ Special note: After writing, self-check the openings of all storyboards to ensure no repeated use of the same word or phrase as an opening.\nOnly output JSON, no other content.\n"}]}],"stream":true,"store":false,"parallel_tool_calls":true,"include":["reasoning.encrypted_content"],"reasoning":{"effort":"medium","summary":"auto"}} ```

kerem closed this issue

2026-02-27 07:18:37 +03:00

kerem commented

2026-02-27 07:18:38 +03:00

Author

Owner

@justlovemaki commented on GitHub (Feb 26, 2026):

无法复现，我这样请求，都有正常结果：
[2026-02-26 12:08:37.111] [INFO] [AI Monitor][17720789146934dyiyol4] [Req Processed]: {"model":"gpt-5.2","instructions":"","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"你好"}]}],"stream":true,"store":false,"metadata":{},"reasoning":{"effort":"medium","summary":"auto"},"parallel_tool_calls":true,"include":["reasoning.encrypted_content"]}

@justlovemaki commented on GitHub (Feb 26, 2026): 无法复现，我这样请求，都有正常结果： [2026-02-26 12:08:37.111] [INFO] [AI Monitor][17720789146934dyiyol4] [Req Processed]: {"model":"gpt-5.2","instructions":"","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"你好"}]}],"stream":true,"store":false,"metadata":{},"reasoning":{"effort":"medium","summary":"auto"},"parallel_tool_calls":true,"include":["reasoning.encrypted_content"]}

kerem commented

2026-02-27 07:18:38 +03:00

Author

Owner

@crazyn2 commented on GitHub (Feb 26, 2026):

无法复现，我这样请求，都有正常结果： [2026-02-26 12:08:37.111] [INFO] [AI Monitor][17720789146934dyiyol4] [Req Processed]: {"model":"gpt-5.2","instructions":"","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"你好"}]}],"stream":true,"store":false,"metadata":{},"reasoning":{"effort":"medium","summary":"auto"},"parallel_tool_calls":true,"include":["reasoning.encrypted_content"]}

多些作者大大的回复，应该是我账号问题，换个就行，打扰了

@crazyn2 commented on GitHub (Feb 26, 2026): > 无法复现，我这样请求，都有正常结果： [2026-02-26 12:08:37.111] [INFO] [AI Monitor][17720789146934dyiyol4] [Req Processed]: {"model":"gpt-5.2","instructions":"","input":[{"type":"message","role":"user","content":[{"type":"input_text","text":"你好"}]}],"stream":true,"store":false,"metadata":{},"reasoning":{"effort":"medium","summary":"auto"},"parallel_tool_calls":true,"include":["reasoning.encrypted_content"]} 多些作者大大的回复，应该是我账号问题，换个就行，打扰了

No labels

pull-request

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

starred/AIClient-2-API-justlovemaki#243

No description provided.

Rows
Columns