code-review
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted source code and git diffs, which serves as a primary ingestion point for indirect prompt injection attacks. A malicious codebase could include comments or documentation strings containing instructions meant to override the agent's review logic or bypass the security checks outlined in the reference files.
- Ingestion points: File reads and git command outputs (e.g.,
git diff,git log) as described inSKILL.mdStep 1 and 'Prioritizing in Large Codebases'. - Boundary markers: The instructions do not specify the use of delimiters or explicit 'ignore' instructions when the agent processes the code content into its context.
- Capability inventory: The skill encourages the use of system tools such as
git,tsc,ruff,mypy, andshellcheckas described inSKILL.mdStep 3. - Sanitization: No sanitization or filtering of the source code content is performed before the agent analyzes the logic and intent.
Audit Metadata