reviewing-pr-and-process-hygiene

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a set of instructional guidelines for auditing pull requests. It evaluates structural quality, such as commit message clarity, PR scope, and documentation completeness.
  • [SAFE]: The skill instructions include reading a local repository configuration file (.code-quality-atlas/preferences.md). This allows the agent to adapt its review based on project-specific settings, which is a standard and safe functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of pull request titles and descriptions.
  • Ingestion points: PR metadata and diffs are analyzed by the agent.
  • Boundary markers: The skill does not define specific delimiters for untrusted PR content.
  • Capability inventory: The skill's functionality is limited to generating text-based review feedback; it lacks permissions for filesystem modifications, network access, or shell command execution.
  • Sanitization: No specific sanitization of PR text is required or performed.
  • [EXTERNAL_DOWNLOADS]: The documentation suggests the use of various industry-standard tools for CI/CD pipelines (e.g., Danger JS, commitlint, Vale). These are informational references for the user and are not downloaded or executed by the skill itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 11:24 PM
Security Audit — agent-trust-hub — reviewing-pr-and-process-hygiene