codebase-onboarding
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs local file reconnaissance to identify project structure and conventions. This behavior is consistent with its stated purpose and lacks malicious intent.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted repositories to generate guide artifacts. Ingestion points: Project manifest files and source code. Boundary markers: None specified in the output templates. Capability inventory: Local file reading and writing. Sanitization: No specific validation of codebase content is mentioned.
Audit Metadata