aibdd-pen-to-storybook

To extract Tailwind tokens from a Pencil .pen file, parse the .pen v2.x JSON schema and map the design elements to normalized Tailwind 4 token namespaces. This read-only process isolates design tokens for downstream Storybook generation without writing files.

Detecting component candidates from a Pencil design involves applying heuristic component detection to a selected screen subtree. This generates a stable component table with derived props and observed story states, providing structured inputs for Storybook planning.

No, you cannot process multiple Pencil screens at once. The extraction is screen-scoped, requiring you to select exactly one top-level screen subtree per run to maintain clean component boundaries and produce an accurate component table.

No, the pen-to-storybook adapter does not modify or scaffold project files. It performs read-only parsing of the .pen JSON to extract Tailwind 4 tokens and component candidates, strictly returning data to the caller without writing anything to disk.

Pencil design extraction fails when the input schema or version is invalid. The adapter validates the .pen v2.x schema upfront and returns structured failure kinds and messages to the caller, ensuring safe dispatch during the ingestion process.

You need to convert a .pen file into Tailwind tokens and component candidates when you have a frozen Pencil design and want to transition to Storybook authoring. This adapter-only ingestion normalizes design data during the discovery and planning phase.