review-pr
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a standard code review process using local development tools like git and internal knowledge graph queries. It does not perform unauthorized network access, exfiltrate data, or attempt to escalate privileges.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code from pull requests, representing a potential surface for indirect prompt injection if malicious instructions are embedded in the code.
- Ingestion points: Pull request diffs and full source code files identified for deep-dive analysis (SKILL.md, Steps 1, 3, and 5).
- Boundary markers: The skill specifies a structured Markdown template for the review output, which provides a level of separation between analysis and results.
- Capability inventory: The agent utilizes local file reading, git command execution, and code-graph search tools.
- Sanitization: There is no explicit sanitization or filtering of the code content described in the skill instructions.
Audit Metadata