ddd
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from repository artifacts like source code, tests, and schemas to inform the creation of the domain model, creating an attack surface for indirect prompt injection.
- Ingestion points: Ingests content from DOMAIN.md, source code, tests, schemas, interfaces, and documentation (as specified in Workflow Step 1).
- Boundary markers: The skill does not define specific markers or delimiters to separate ingested file content from its internal instructions.
- Capability inventory: The skill has read access to repository files and write access to the DOMAIN.md file at the repository root.
- Sanitization: No explicit sanitization or filtering of the ingested repository content is mentioned. This risk is mitigated by the operating contract which requires the user to review and confirm all document changes before they are finalized.
Audit Metadata