supa-review
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data, specifically git diffs and repository files, as its primary function for code review. There is an inherent surface for indirect prompt injection where instructions embedded in the code under review could attempt to influence the agent's behavior.
- Ingestion points: Git status and diff outputs, repository source files, and external documentation retrieved via research tools.
- Boundary markers: The instructions do not specify the use of delimiters (such as xml tags or markdown blocks) to isolate untrusted code content from the agent's instructions.
- Capability inventory: The skill is restricted to read-only operations (file reads and network research) unless the user explicitly requests fixes, and it does not define capabilities for subprocess execution, file writing, or credential access.
- Sanitization: No explicit sanitization or filtering logic is provided for the ingested code content.
Audit Metadata