review-main
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill executes standard Git commands (
git merge-base,git diff) to identify changes between branches. These are restricted to local repository operations and are consistent with the skill's stated purpose. - [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8) because it ingests and processes untrusted data from
git diffoutput. - Ingestion points: The content of files being reviewed via
git diffis fed directly into the agent's context. - Boundary markers: None specified in the workflow to separate code content from instructions.
- Capability inventory: The agent performs analysis and reporting based on the ingested code; while it doesn't have destructive capabilities listed, the review output could be biased by malicious comments in the code.
- Sanitization: No sanitization or filtering of the diff content is performed before analysis.
Audit Metadata