receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill's core functionality involves processing and implementing suggestions from external reviewers, which establishes an indirect prompt injection surface. A malicious actor could provide feedback containing instructions for the agent to bypass checks or perform unauthorized code modifications.
- Ingestion points: The agent is instructed to read and process feedback from both human partners and external reviewers in SKILL.md.
- Boundary markers: Absent. The skill does not define or use delimiters (like XML tags) to isolate untrusted feedback from the system prompt.
- Capability inventory: The agent has access to shell tools (grep), the GitHub CLI (gh api), and the ability to modify project files.
- Sanitization: Absent. There is no mechanism described for validating or sanitizing the content of the review feedback before it is evaluated for implementation.
Audit Metadata