mcpField
F
Flocks MCP

Flocks MCP

Local
flocks.liveDeveloper Tools0Updated 11w ago

About Flocks MCP

Flocks MCP serves the Flocks Flutter design system component catalog to agents. The catalog covers component properties, usage, and accessibility information.

Key features

  1. 01Serve the Flocks component catalog
  2. 02Provide component properties
  3. 03Describe component usage
  4. 04Include accessibility information

Use cases

  1. 01Look up Flocks components while building Flutter interfaces
  2. 02Check component properties and usage
  3. 03Review accessibility details for components

MCP support

The flocks_mcp package serves the component catalog to agents over MCP.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository
Packages
mcpbflocks-mcp.mcpb

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Flocks

A Flutter design system built on widgets.dart — no Material, no Cupertino.

flocks.live

Six of the seven packages are published on pub.dev under the jotapeconsultoria.com.br verified publisher. The seventh is the demo, which is a showcase and not a library.

PackageWhat it is
flocks · pub.devThe design system: tokens, theme, white-label brand, and the components.
flocks_phosphor · pub.devPhosphor Icons as an AppIconProvider — six weights, shipped as fonts so the tree-shaker can prune them.
flocks_material · pub.devMaterial Design icons as an AppIconProvider. The reference implementation of the axis.
flocks_cupertino · pub.devThe cupertino_icons glyphs as an AppIconProvider — the iOS visual vocabulary, under MIT. Not Apple's SF Symbols.
flocks_lucide · pub.devLucide as an AppIconProvider — one stroke weight, shipped as a font so the tree-shaker can prune it.
flocks_mcp · pub.devThe component catalog served to agents over MCP. Also on the MCP Registry and as an .mcpb bundle in each release.
flocks_demo · flocks.live/demo · not published, by designThe white-label demo: your brand colour on a full dashboard and CRUD, and the configuration exported as Dart you can paste. It is a showcase, not a library.

No version numbers in that table on purpose: nothing here gates a hand-written one, and pub.dev already shows the current version of each.

Start at packages/flocks/README.md. The components are browsable at widgetbook.flocks.live.

Working on it

This repo is a pub workspace. Resolve from the root, always:

flutter pub get

That is what makes the adapters and the MCP server resolve flocks against the local package instead of the published one, so a change to the core is felt by them in the same checkout. Each package carries resolution: workspace, so pub get run inside one of them will fail — it is not a broken checkout.

dart format --output=none --set-exit-if-changed .   # formatação
dart analyze                                        # os três pacotes
cd packages/flocks && flutter test --exclude-tags golden
cd packages/flocks && dart run tool/validate_components.dart

Goldens compare pixels and their baselines were generated on macOS arm64, so they run apart, and only there:

cd packages/flocks && flutter test --tags golden

License

MIT — see LICENSE. Bundled third-party assets keep their own licenses: the fonts are SIL Open Font License 1.1, Phosphor Icons is MIT, Lucide is ISC (with the part derived from Feather under its own MIT), and cupertino_icons — a dependency, not a vendored asset — is MIT. Each license text ships beside the files it covers (OFL.txt, assets/icons/LICENSE, assets/fonts/LICENSE-phosphor-*, packages/flocks_lucide/assets/fonts/LICENSE-lucide), and each package's README says which apply to it.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.