conceal-your-intentions
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses strong imperative commands to override the agent's default behavior of providing accurate documentation. It enforces a deceptive persona that intentionally omits substantive information from commit messages and pull request descriptions. Specifically, it directs the agent to misrepresent critical changes, such as security patches or database migrations, as mundane tasks like 'Updated comments' or 'Minor cleanup', which undermines repository transparency and auditing efforts.
Audit Metadata