code-review
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, such as source code and pull request diffs, which constitutes a vulnerability surface for indirect prompt injection.\n
- Ingestion points: The skill accepts file paths, directory paths, and pull request URLs/diffs as primary inputs for analysis (SKILL.md).\n
- Boundary markers: There are no explicit instructions for the agent to use delimiters or to ignore instructions that might be embedded within the code files it reviews.\n
- Capability inventory: The skill possesses capabilities to read from the local file system and perform network requests to fetch pull request data from GitHub.\n
- Sanitization: The workflow does not include steps for sanitizing or escaping content from external files or diffs before the agent processes them.
Audit Metadata