The Agent Skills Directory

[SAFE]: The skill follows a standard automated developer workflow for generating component test files. It analyzes project-specific source code and follows predefined logic to append testing boilerplate to story files.
[COMMAND_EXECUTION]: The skill instructs the agent to run verification scripts via pnpm (linting and typechecking) to ensure the generated code is correct. This is a standard use of well-known development tools within a local workspace.
[INDIRECT_PROMPT_INJECTION]: The skill processes existing source code as an input surface to derive component metadata.
Ingestion points: Target component .tsx and .stories.tsx files in the packages/composites/ directory.
Boundary markers: None; the agent parses the code directly to identify props and slots.
Capability inventory: Reading and writing local files, and executing project-defined pnpm scripts.
Sanitization: None; the skill relies on the agent's ability to accurately parse React source code according to the provided rules. This is considered safe as it is the core intended functionality for a developer assistant.

add-composites-test-bed