no-comments
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to read an external configuration file (
../saltmode/references/host-adapter.md) to set 'authority limits' and map tool permissions. This is an attempt to override or influence the agent's core instructions with external content. - [PROMPT_INJECTION]: The skill mandates the removal of code comments, specifically targeting 'Correctness or safety suppressions' as 'MUST KILL' items. It instructs the agent to err on the side of deletion if a comment's purpose is 'ambiguous,' which could lead to the unintended removal of vital security markers, safety documentation, or linter suppressions intended to prevent vulnerabilities.
Audit Metadata