scratch
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes arbitrary user-provided text via the
/scratchcommand and stores it using task management tools. This introduces an attack surface where a user could inject malicious instructions into the scratchpad content. The risk is classified as low because the skill is restricted to session-scoped in-memory storage and does not have network or file system access.
Audit Metadata