code-review
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes Git CLI commands including
git diff,git log, andgit rev-parseto retrieve the code changes and commit history required for the review. These operations use a user-provided reference point (e.g., a commit SHA or branch name). - [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it ingests and processes untrusted data from the repository's history and external specification files.
- Ingestion points: The agent reads output from
git diff,git log, and fetches contents from files located indocs/,specs/, or.scratch/directories. - Boundary markers: The instructions do not specify explicit delimiters or markers to isolate the untrusted code/spec content from the sub-agent instructions.
- Capability inventory: The skill utilizes the
Agenttool to delegate tasks to parallel sub-agents for specialized analysis. - Sanitization: There is no evidence of sanitization or filtering of the code diffs or specification text before they are passed to the sub-agents.
Audit Metadata