The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard shell commands to inspect the project environment and perform code quality checks. It reads configuration files and runs built-in developer tools.
Evidence: Use of cat and ls for file inspection; execution of npx tsc, npx eslint, and npm run dev to verify the generated source code.
[EXTERNAL_DOWNLOADS]: The skill suggests installing necessary frontend dependencies from official package registries if the project environment is not fully configured.
Evidence: Instructions to run npm create vite and npm install for framework and utility setup in Step 2.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests content from local markdown files within the .design/ directory to guide its code generation process.
Ingestion points: Design specification files such as DESIGN.md, CODE_PATTERNS.md, COMPONENTS.md, PAGES.md, and ICONS.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are implemented.
Capability inventory: The skill has the capability to write files to the src/ directory and execute shell commands (npx, npm).
Sanitization: There is no evidence of validation or sanitization of the input design specs before they are used to generate executable code.

design2code