The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run pnpm check and pnpm build in the project root to verify the integrity of the generated components. These commands are standard development practices for linting and build verification.
[INDIRECT_PROMPT_INJECTION]: The workflow involves reading existing files in packages/smoothui/components/ to research patterns. While this is a data ingestion surface, it is a standard requirement for code-generation skills to maintain consistency within a repository.
Ingestion points: Reads existing component files in packages/smoothui/components/ during the research phase.
Boundary markers: None specified for the file-reading operations.
Capability inventory: The skill has capabilities to write files (components, documentation, configuration) and execute shell commands (pnpm).
Sanitization: No explicit sanitization of input files is mentioned; the skill relies on the agent's interpretation of existing code patterns.

smoothui-component-craft