review-recipient
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing external, potentially untrusted feedback from reviewers, which creates a surface for indirect prompt injection attacks where malicious instructions could be embedded in code review comments.
- Ingestion points: Code review feedback from "External Reviewers" as described in the Source-Specific Handling section.
- Boundary markers: The skill provides logical boundaries by instructing the agent to "READ: Complete feedback without reacting" and "VERIFY: Check against codebase reality" before any implementation occurs.
- Capability inventory: The agent is expected to use file-reading tools (e.g.,
grep) and perform file-writing operations to implement suggested code improvements. - Sanitization: The skill mandates technical verification, skepticism towards external suggestions, and strict adherence to YAGNI (You Aren't Gonna Need It) principles, which serves as a manual sanitization process to prevent the execution of malicious or unnecessary instructions.
Audit Metadata