Rank, 14 days: 170 on the Codex sheet
30 people reported
All time, since Sep 16, including 5 thumbs-up on GitHub.
GPT-6 prompts rejected
“Invalid prompt: your prompt was flagged as potentially violating our usage policy.”
Is this your bug? What gets counted gets fixed. Pile on.
Paste this into your agent. It files your report, with your error and version.
I hit this bug: rapsheet.dev/sheets/codex/issues/9354/file Follow it to file my report. Show me the text before you send.
No install. No account. Public and anonymous. The link asks your agent for one report, nothing else. What it sends
Simple prompts rejected with invalid_prompt on GPT-6 models. Even "Hello" fails as "flagged as potentially violating our usage policy" in Codex desktop app and CLI, while GPT-5.6 Sol works.
Is this your bug?
File my report- Where
- Codex desktop app
- Error text
Invalid prompt: your prompt was flagged as potentially violating our usage policy. Please try again with a different prompt: https://platform.openai.com/docs/guides/reasoning#advice-on-prompting- Versions
- 0.157.1 to 26.1013.40934
- Platforms
- Linux, macOS, Windows
- How bad
- Breaks a normal workflow. Some reports say it blocks them completely.
Not quite your bug?
- Models rejected as unsupported2 people
- Sol model rejected on ChatGPT sign-in19 people
- Prompts rejected as invalid_prompt3 people
Original reports26, newest first
- Safety filter regression in v6/v6.1 – benign prompts blocked (v5.6 unaffected)On GitHub#57043
- Invalid promptOn GitHub#55766
- GPT-6 family models all return invalid_prompt in Codex — request for help (my research is being blocked)On GitHub#55340
- Every prompt flagged as "Invalid prompt: potentially violating usage policy" — false positive on all inputs, even "hello"On GitHub#54874
- Every request to the GPT-6 models is currently rejected, including the harmless prompt "haloOn GitHub#53925
- False-positive invalid_prompt for benign local GUI maintenance (gpt-6-astra)On GitHub#53856
- When using the GPT model, it prompts no matter what is being sent, , GPT 5.6 is not functioning properlyOn GitHub#53580
- GPT-6.1 Sol rejects benign prompts as invalid_prompt while GPT-5.6 succeedsOn GitHub#52450
- [Windows] GPT-6 Astra and GPT-6.1 Sol reject "hello" with invalid_prompt; GPT-5.6 Terra respondsOn GitHub#52024
- your prompt was flagged as potentially violating our usage policyOn GitHub#51939
Show all 26Show only the newest 10
- GPT-6 models reject all harmless prompts as policy violationsOn GitHub#51912
- Codex task errorOn GitHub#51242
- [Windows][Codex Desktop] GPT-6 models (Astra/Sol) reject every prompt with invalid_prompt, GPT-5.6 works fineOn GitHub#50657
- False-positive invalid_prompt in Codex App on GPT-6 Sol / GPT-6.1 Sol / GPT-6 Astra while GPT-5.6 Sol worksOn GitHub#50605
- Invalid prompt: your prompt was flagged as potentially violating our usage policy. for GPT6On GitHub#49440
- All GPT-6 models available to my account are unusable due to invalid_promptOn GitHub#49213
- Invalid prompt ResponseOn GitHub#49120
- GPT-6 Sol/Luna/Astra falsely reject benign prompts with Invalid prompt safety errorOn GitHub#48817
- BUG：GPT-6 Astra flags every harmless prompt as a policy violation in Codex Desktop and WorkOn GitHub#48713
- [Windows][Codex Desktop] GPT-6 Astra rejects a fresh task with invalid_promptOn GitHub#47551
- [Codex Desktop] GPT-5.6 Sol and GPT-6 Astra reject harmless prompts with invalid_promptOn GitHub#47041
- [Windows Desktop] GPT-6 Astra and GPT-5.6 Sol reject harmless prompts with invalid_prompt while Codex Web worksOn GitHub#46756
- False positive policy flag on harmless prompts, automatic model downgradeOn GitHub#46753
- [Windows] GPT-6 Astra rejects every prompt with "Invalid prompt"On GitHub#46082
- gpt-6-astra rejects benign arithmetic in isolated CLI 0.154.0; gpt-5.6-sol succeedsOn GitHub#45993
- Model restrictionsOn GitHub#45963