review-quick
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes several standard git commands (rev-parse, log, diff, status, ls-files) in SKILL.md Step 1 to collect project changes. These commands are executed locally and are strictly used to gather context for the review agents.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. Ingestion points: The skill reads output from git diff and the full contents of untracked files in SKILL.md Steps 1 and 2. Boundary markers: The instructions in Step 3 do not specify delimiters or instructions for sub-agents to ignore embedded commands in the reviewed content. Capability inventory: The skill performs shell execution (git commands) and file reading. Sanitization: No sanitization is applied to the ingested content before dispatching to sub-agents. This represents a standard vulnerability surface for automated review tools and is not considered malicious here.
Audit Metadata