audit-code-review

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a comprehensive code review workflow with clear phases for observation, interpretation, and classification. It correctly prioritizes security findings (like IDOR and SQL injection) as blocking issues.
  • [COMMAND_EXECUTION]: The skill uses standard git commands (git log, git diff) to understand the context of changes. These are typical for developer-focused tools and do not involve unsanitized user input.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes firecrawl_search to research current best practices. This is a well-known service used for information gathering and does not involve executing remote code or downloading untrusted packages.
  • [DATA_EXPOSURE]: The skill integrates with Sentry (sentry:search_issues) to check for production impact. This is a standard observability practice and is performed within the developer's authorized environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 05:55 AM
Security Audit — agent-trust-hub — audit-code-review