agency-design-system
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted content from project files, creating a surface for indirect prompt injection. This is assessed as safe because the skill's capabilities are restricted and its output is limited to design documentation.
- Ingestion points:
copy.mdand.agency/context/visual-identity.md. - Boundary markers: None identified in the provided instructions.
- Capability inventory: The skill is restricted to read-only file system tools (
Read,Grep,Glob). - Sanitization: No explicit sanitization or validation of ingested text is performed.
Audit Metadata