skills/uinaf/agents/review-gang/Gen Agent Trust Hub

review-gang

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted code and documentation from external repositories. This creates a surface for Indirect Prompt Injection, where malicious code could attempt to manipulate the review outcome.
  • Ingestion points: Reads repository guidance (e.g., AGENTS.md, CLAUDE.md) and code diffs as described in SKILL.md.
  • Boundary markers: Lacks structural delimiters to isolate untrusted content within the reviewer subagent prompts.
  • Capability inventory: Can execute git commands and project-specific tests (e.g., pnpm test).
  • Sanitization: No sanitization is performed on external content before it is processed.
  • [REMOTE_CODE_EXECUTION]: The skill encourages running the project's own tests (e.g., pnpm test) to verify behavior. Executing tests from an unvetted repository is a security risk, although it is a primary function of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 03:25 PM
Security Audit — agent-trust-hub — review-gang