[Bug] Unrelated gambling/adult terms appear in Codex auto-generated titles and agent output
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.4k
- PR merge metrics
- PR metrics pending
Description
What happened
Codex is sometimes inserting unrelated gambling/adult/spam-like Chinese text into otherwise normal research tasks.
I have seen this in two places:
-
Auto-generated conversation titles
完成立场测试 国产成人精科研记忆原则 天空彩票网
The added suffixes are completely unrelated to the conversations.
-
Agent work/reasoning output
- Similar unrelated gambling/adult text occasionally appears while Codex is working on research tasks.
- These terms were not present in my prompt or project context.
Expected behavior
Codex should not insert unrelated gambling/adult/spam text into titles or agent output.
Reproduction
The problem is intermittent, so I do not yet have a deterministic reproduction.
I have a screenshot showing the malformed auto-generated titles and can provide more examples privately if needed.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing the reported malformed auto-generated titles, the agent work/reasoning output examples, and the available screenshot. Investigate the title-generation and agent-output paths for how unrelated text could be introduced; done means neither path inserts unrelated gambling, adult, or spam-like text, with coverage for the reported cases.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- ai
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100