implement
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's
/implementcommand processes user-supplied feature descriptions without explicit boundary markers or sanitization. 1. Ingestion points: The [feature-description] argument in SKILL.md. 2. Boundary markers: None defined to isolate user input from agent instructions. 3. Capability inventory: The skill allows for file modification and access to MCP servers for code generation. 4. Sanitization: No validation or filtering of the input is specified. - [NO_CODE]: The skill provides instructional documentation and lacks executable scripts, which limits the direct execution risk.
Audit Metadata