design-review
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided design documents and related project files (e.g., design/gdd/ files) and interpolates this content into prompts for subagents spawned via the Task tool.
- Ingestion points: Primary design document provided as an argument and secondary documents discovered via Glob in Phase 1.
- Boundary markers: The skill lacks explicit boundary markers or 'ignore embedded instructions' delimiters when passing document content to specialist subagents in Phase 3b.
- Capability inventory: The skill has access to Read, Glob, Grep, Write, Edit, and Task tools, which enables the processing of external content and the spawning of independent sub-sessions.
- Sanitization: No validation or sanitization of the design document content is performed before it is used to prompt subagents.
Audit Metadata