The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use shell commands (open on macOS, xdg-open on Linux, and start on Windows) to open Pinterest search URLs in the user's default browser. This is used for visual reference discovery.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files (e.g., molten-docs/brand/brand.md) and user-provided images to influence the generation of design tokens and search terms. This creates a surface where content in those files could influence the agent's output.
Ingestion points: Reads brand strategy from molten-docs/brand/brand.md, legacy brand.md, and processes user-uploaded images/descriptions.
Boundary markers: The instructions do not specify explicit boundary markers when interpolating content from the brand brief into the design system generation prompt.
Capability inventory: The skill writes files to the local file system (molten-docs/design/design.md, molten-docs/design/example.html) and executes shell commands to open browser tabs.
Sanitization: The skill mandates URL-encoding for external search terms and includes a confirmation gate where the user must approve the interpretation of reference images before they are applied to tokens.

molten-design