find-peers
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to discover and analyze untrusted external content from public repositories on GitHub and Sourcegraph. This ingestion of third-party source code and manifests creates a surface for indirect prompt injection.
- Ingestion points: External source code, package.json files, and lockfiles are fetched and analyzed from third-party repositories found via search queries.
- Boundary markers: The skill does not define explicit delimiters or boundary instructions for the external content being processed.
- Capability inventory: The skill possesses capabilities to execute local shell commands (pnpm) and write analysis results to files in the .brain/ directory.
- Sanitization: The skill relies on natural language instructions to 'verify correctness before copying' and 'read those files' to ensure safety, rather than programmatic isolation.
- [COMMAND_EXECUTION]: The skill executes local pnpm commands to perform discovery and refresh tasks.
- Evidence: pnpm find-peers and pnpm find-peers --min 1 in SKILL.md.
Audit Metadata