compound-docs
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill mandates that the agent 'ALWAYS search solutions before investigating,' creating a surface where malicious instructions could be embedded in documentation files stored in
.claude/solutions/. - Ingestion points: Solution files located in the
.claude/solutions/subdirectories as defined in SKILL.md. - Boundary markers: The agent is not provided with delimiters or instructions to treat documentation content as untrusted.
- Capability inventory: The agent can search, read, and write files, and uses the documentation to inform its investigation and planning logic.
- Sanitization: Content is not sanitized or filtered for instructions before being processed by the agent.
Audit Metadata