cdk-documentation-review
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Trusted External References: The skill references guidelines and documentation from the official AWS CDK repository on GitHub. These references are used to provide the agent with authoritative criteria for its review process.
- Indirect Prompt Injection Surface: The skill is designed to ingest and analyze external data, specifically Pull Request diffs. This ingestion represents a potential indirect prompt injection surface. The analysis chain involves: (1) Ingestion point: Pull Request diffs; (2) Boundary markers: No specific markers mentioned; (3) Capability inventory: No file-write, network, or subprocess capabilities identified; (4) Sanitization: None specified. The risk is minimized by the lack of high-privilege tool access.
Audit Metadata