social-youtube-thumbnail
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the user's message directly to construct a design brief for the open-design tool. This lacks input sanitization or explicit boundary markers, potentially allowing a user to include instructions that could influence the agent's behavior during the design process.
- Ingestion points: SKILL.md (Brief interpolation from user message)
- Boundary markers: None specified
- Capability inventory: list_open_design_design_systems, recommend_open_design_design_systems, prepare_open_design_brief, register_open_design_artifact, and local file-editing tools
- Sanitization: None specified
- [DYNAMIC_EXECUTION]: The agent is instructed to author files based on instructions returned from the
prepare_open_design_brieftool. This relies on the integrity of the design system's output to guide local file modifications.
Audit Metadata