skill-review
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process the content of external files including SKILL.md, documentation in the references/ folder, and various scripts associated with the skill being reviewed. These files represent untrusted data that could contain malicious instructions designed to subvert the agent's behavior during the review process.
- Ingestion points: Reads SKILL.md, files in the references/ directory, and scripts referenced within the skill's instructions.
- Boundary markers: The skill does not define or use explicit delimiters or "ignore previous instructions" guards for the content it ingests from external files.
- Capability inventory: The skill utilizes filesystem tools such as Read, Grep, and Glob, and has access to the Bash tool for script inspection and general operations.
- Sanitization: No sanitization, escaping, or filtering of the ingested external content is performed before the agent processes and analyzes it.
Audit Metadata