design-variations

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user requests and project data to generate design variations, creating a potential surface for embedded instructions to influence the agent.
  • Ingestion points: User requests, project baselines, components, and domain language described in Workflow Step 1 and Step 2 of SKILL.md.
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the input data.
  • Capability inventory: The agent writes to a local HTML file (assets/variations.html) and eventually promotes a winner to the production stack (Workflow Step 7).
  • Sanitization: The skill does not explicitly mention sanitizing user input before including it in the generated artifact.
  • [DYNAMIC_EXECUTION]: The skill involves generating executable HTML and JavaScript code for UI prototyping.
  • Evidence: Workflow Step 4 in SKILL.md instructs the agent to build a disposable browser artifact by populating the assets/variations.html template with generated code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 09:17 AM