repo-maintainer

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs various shell operations to audit repository health, including git state inspection (git status, git diff, git remote -v, git log) and the execution of project-specific scripts like npm run merge:batch and FAF-specific commands.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by design. It explicitly instructs the agent to read and delegate to policy files found within the target repository, creating a risk that malicious instructions in those files could hijack the session.
  • Ingestion points: root and nested AGENTS.md, contributor guidance, maintainer docs, and release instructions from the target repository.
  • Boundary markers: None; the skill does not use specific delimiters or warnings to ignore embedded instructions within the files it reads.
  • Capability inventory: The skill has capabilities for file system read/write, shell command execution (Git and NPM), branch management, and pull request creation.
  • Sanitization: There are no instructions for sanitizing or validating the content of the policy files before following their directives.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 12:46 PM
Security Audit — agent-trust-hub — repo-maintainer