code-review
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external codebases, including pull request descriptions, comments, and diffs. An attacker could embed instructions within this content to attempt to influence the agent's review or fixing behavior.
- Ingestion points:
SKILL.md(Step 1) instructions specify fetching platform diffs, PR descriptions, and comments. - Boundary markers: The skill does not define specific delimiters or instructions to the agent to disregard natural language commands found within the data being reviewed.
- Capability inventory: The skill utilizes file read/write operations and git command execution to perform its audit and fix loops.
- Sanitization: There is no evidence of filtering or sanitizing ingested content before it is processed by the agent.
Audit Metadata