Design System CLI Workflows for Component Libraries

A design system CLI is most valuable when it becomes the shared pipeline for docs, checks, and AI tooling. Without that shared pipeline, teams end up with stale metadata, mismatched docs, and inconsistent CI behavior.

Fragments uses the CLI as the source of truth for compiling component metadata and enabling downstream workflows.

## Core CLI Workflow for Component Libraries

A component library CLI workflow should make it easy to:

1. initialize metadata/config 2. build compiled component data 3. validate output in CI 4. power docs and AI integrations from the same artifact

The [CLI Reference](/governance/cli) documents each command and option. The compiled output then powers the [MCP Tools](/governance/mcp) and supports review in the [components docs](/components).

## Why This Matters for AI-Assisted UI Work

If AI is part of your workflow, a stable CLI pipeline matters even more. The assistant needs a fresh artifact (`fragments.json`) to query current component APIs and guidance.

## FAQ

### Can I use the CLI without enabling MCP yet?

Yes. The CLI is useful on its own for compiling metadata and validating coverage. MCP is an optional integration layer on top.

### Where should teams start?

Use [Getting Started](/components/install) for the setup flow, then use the [CLI Reference](/governance/cli) as the long-term command reference.

## CTA

Use the [Design System CLI Reference](/governance/cli) to build a repeatable metadata pipeline, then connect [MCP Tools](/governance/mcp) when your team is ready for AI-assisted workflows.