design
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to follow instructions from an external file.
- Ingestion points:
.agents/workflows/design.mdreferenced inSKILL.md. - Boundary markers: Absent. The agent is directed to "Read and follow... step by step" without instructions to ignore potentially malicious embedded commands.
- Capability inventory: The skill file itself contains no code or tool calls, but the design workflow typically implies file system and agentic capabilities.
- Sanitization: Absent. No validation or filtering is performed on the content of the referenced markdown file.
Audit Metadata