The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run npx playwright screenshot to verify the rendered UI layout. This uses a well-known browser automation tool from a trusted vendor for visual testing of local files.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted design data from external Figma URLs and a local design-context file, which could influence agent output. 1. Ingestion points: Figma designs (via MCP) and .agents/design-context.md (both referenced in SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Command execution via npx. 4. Sanitization: Absent; external design tokens and context are used directly in layout generation.

card-design