skills/btraut/skills/review-team/Gen Agent Trust Hub

review-team

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill follows best practices for multi-agent orchestration, including:
  • Read-Only Constraints: Sub-agents are explicitly instructed to be read-only and are prohibited from editing files or making commits, limiting the impact of any potential automated actions.
  • Command Execution: System commands are restricted to local, non-destructive git operations (git log, git show, git status, git diff) used solely for scope discovery and intent analysis.
  • Input Handling: While the skill processes untrusted source code (a surface for indirect prompt injection), it implements structured schema validation (findings-schema.json) and merge-normalization rules to ensure integrity and consistency of the analysis output.
  • Privacy: The skill does not perform external network requests to untrusted domains; external references are limited to well-known standard schema repositories.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 01:13 AM