ml-model-code-review
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from GitHub Pull Requests (identified by the
repositoryandpr_numberconfiguration fields). This creates an attack surface for indirect prompt injection, where malicious instructions embedded in the code or PR comments could influence the agent's behavior during the review process. - Ingestion points: Pull Request code and metadata from the specified GitHub repository.
- Boundary markers: None identified in the prompt templates to separate untrusted data from instructions.
- Capability inventory: The skill is scoped to analysis and reporting within the platform's code review feature; no dangerous execution or unauthenticated network tools are invoked in the provided logic.
- Sanitization: No explicit sanitization or filtering of PR content is performed before the content is processed by the AI.
Audit Metadata