requesting-code-review

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The code-reviewer.md subagent template ingests code diffs from git diff. Without explicit boundary markers or instructions to ignore embedded commands, the subagent could be manipulated by adversarial text or instructions contained within the code being reviewed.
  • Ingestion points: code-reviewer.md (via git diff output from provided commit range).
  • Boundary markers: Absent. The diff content is provided directly to the subagent without specific delimiters to distinguish between code and potential malicious instructions.
  • Capability inventory: The subagent performs code analysis, quality assessment, and production readiness evaluation.
  • Sanitization: Absent.
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands including git rev-parse, git log, and git diff to identify commit ranges and extract code changes for review. These are standard, necessary operations for the skill's stated purpose of facilitating code reviews.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 12:36 AM