agent-first-repo

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill describes an autonomous workflow for 'Entropy Management' in 'references/entropy-management.md' where an agent scans the codebase for violations and opens refactoring PRs. This creates an indirect prompt injection surface.
  • Ingestion points: The workflow involves agents reading the entire codebase, including files that may contain untrusted or externally-sourced content.
  • Boundary markers: The skill does not define specific boundary markers or instructions to help the agent distinguish between data (code) and instructions during its automated scans.
  • Capability inventory: The recommended workflow assumes the agent has permissions to modify the repository via 'targeted refactoring PRs' and file updates to 'QUALITY_SCORE.md'.
  • Sanitization: There is no mention of sanitizing or escaping content read from files before processing it as part of the automated audit.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 09:33 PM