code-review

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses user-supplied references (e.g., branch names, commit SHAs) directly in shell commands such as git diff <fixed-point>...HEAD and git rev-parse <fixed-point>. If the execution environment does not properly sanitize these arguments, it could be vulnerable to command injection.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from repository content, including commit messages, code diffs, and external spec files. This content is interpolated into sub-agent prompts for the 'Standards', 'Spec', and 'Frontend' lanes without boundary markers or sanitization.
  • Ingestion points: Git diff output, commit logs, untracked files, and spec files (e.g., docs/, specs/).
  • Boundary markers: Absent in sub-agent prompts.
  • Capability inventory: Shell execution (git commands), file system read access, and tool invocation (/frontend-review, /setup-universal-agent-skills).
  • Sanitization: No evidence of sanitization or escaping of external content before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 06:57 AM
Security Audit — agent-trust-hub — code-review