w-council
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the codebase, including README files, scripts, and package metadata, which creates a surface for indirect prompt injection.
- Ingestion points: The skill instructs the agent to 'Inspect the area yourself first: owning packages, README/scripts, and current behavior' in SKILL.md.
- Boundary markers: There are no instructions for the agent to use delimiters or to ignore instructions found within the analyzed files.
- Capability inventory: The skill can spawn multiple explorer agents and hand off to other skills (w-build, w-plan).
- Sanitization: No sanitization or validation of the codebase content is performed before processing.
Audit Metadata