code-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes content from pull requests, which is untrusted external data. This creates a potential surface where embedded instructions in the code could attempt to influence the agent.
- Ingestion points: External data enters via pull request files and metadata processed by subagents.
- Boundary markers: The skill does not define specific delimiters to isolate untrusted data from its core instructions.
- Capability inventory: It has the ability to apply labels to GitHub pull requests and output review findings.
- Sanitization: The instructions do not specify sanitization or validation of the ingested code content.
Audit Metadata