frontend-design
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an ingestion surface by instructing the agent to prioritize content from an external file (
DESIGN.md).\n - Ingestion points: The agent is instructed to read
DESIGN.mdand the repository's existing styling system.\n - Boundary markers: Absent; the skill explicitly directs the agent to let the external file 'win' and override its own instructions, potentially prioritizing malicious instructions if present in the data.\n
- Capability inventory: The agent has the capability to modify and create web UI components, layouts, styling, and motion across the project.\n
- Sanitization: No sanitization or validation of the external file content is specified before the agent adopts its configuration.
Audit Metadata