code-review
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes untrusted external feedback, which is an injection surface.\n
- Ingestion points:
references/code-review-reception.mdhandles comments from external reviewers.\n - Boundary markers: Absent; no delimiters or instructions to ignore embedded commands are present.\n
- Capability inventory: Local git command execution (
git rev-parse,git log) and task tool dispatching to subagents.\n - Sanitization: Absent; feedback content is processed without validation.\n- [Command Execution] (SAFE): The skill correctly uses git and grep for repository state analysis and verification.
Audit Metadata