project-discovery
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill includes explicit instructions for the agent to identify environment variables and secrets without exposing their actual values, which is a positive security practice.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from untrusted repository files, which represents an attack surface for indirect prompt injection.
- Ingestion points: The agent is instructed in SKILL.md to read manifest files (e.g., package.json, Cargo.toml), configuration files, and source code files.
- Boundary markers: The prompt does not provide specific delimiters or instructions to ignore potential commands embedded in project files.
- Capability inventory: The skill is intended for use prior to code modification tasks, meaning the agent context typically involves filesystem write permissions.
- Sanitization: No explicit sanitization or validation of the retrieved text is mentioned in the discovery instructions.
Audit Metadata