setup
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements an interactive onboarding flow that gathers user information and stores it in a long-term memory file.
- Ingestion points: Step 2 collects free-text answers from the user regarding their name, role, and interests. Step 4 reads configuration keys from a local
.mcp.jsonfile. - Boundary markers: No boundary markers, delimiters, or framing instructions are used to isolate the user-provided data when it is written to the profile file or read back into context.
- Capability inventory: The skill is granted extensive permissions including
Bashfor command execution, andRead,Write, andEditfor file system management. - Sanitization: The skill does not perform any validation, escaping, or filtering of the user-provided strings before they are appended to
memory/user-profile.md.
Audit Metadata