False Positive - Flagged For Cyber Security Risk
Resolved 💬 3 comments Opened Feb 18, 2026 by SovereignBuilder Closed Feb 20, 2026
💡 Likely answer: A maintainer (github-actions[bot], contributor)
responded on this thread — see the highlighted reply below.
When asking codex to help me build a skill / tool that would allow him to become a detective for finding recent activity by AI agents on my PC, my account was flagged for potential high-risk cyber security activity.
This is a first-time occurence for me. I only needed help finding what recent directories any AI have been working from within because I forget constantly and lose valuable context and can't recover/resume prior chat sessions.
I have only just yesterday learnt how to build skills and thought this would be a perfect example of a repeatable skill codex could perform.
Please advise if there are any restrictions in place or if there is anything I need to do to resolve this.
Warm Regards,
Jing
3 Comments
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
I really think the codex team and this cyber verify team are in 2 separate valleys. What a wonder sight of failed project management . they put the system live before announcing it. No visibility what so ever on it in the ACTUAL client that uses the model. Silently kicking people off a model, then now us having to deal with consequences instead of A/B testing or stress testing this.
Thanks for the bug report. We're continuing to tune our cyber safety classifiers to make them more accurate. You can read more about this here. If you see this again in the future and believe it's a false positive (i.e. you aren't doing anything related to cybersecurity), please use the
/feedbackslash command and choose "Safety Check" to report it. That will help us further tune our checks.