review-panel
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface
- Ingestion points: The skill ingests untrusted data in the form of code diffs and file contents via
git diffandgh pr diffduring Step 1. - Boundary markers: The dispatch prompt for subagents uses structured labels (Scope, Changed files, Diff) but lacks robust "ignore embedded instructions" warnings for the content of the hunks.
- Capability inventory: The orchestrator utilizes the
Agenttool to dispatch specialized reviewers (sr-reviewer, sa-reviewer, qa-reviewer, etc.). - Sanitization: Diff content is passed directly to subagents without sanitization, which is standard for code review tools but inherently carries the risk of processing malicious instructions embedded in comments or code within the diff.
Audit Metadata