MCP Server Setup

The fluttersdk artisan CLI ships with a built-in Model Context Protocol (MCP) server that exposes your running Flutter app to AI coding assistants. The server speaks stdio JSON-RPC and surfaces tools contributed by every registered ArtisanServiceProvider (substrate commands plus Dusk, Telescope, and any custom plugin).

This page documents the per-client install matrix: the exact configuration snippet each MCP-compatible client needs to spawn dart run fluttersdk_artisan:mcp against your project. The artisan MCP server is stdio-only in V1 (no remote HTTP endpoint); every client below launches the Dart process locally with the project directory as the working directory.

Without artisan MCP

LLM agents working against a running Flutter app are blind without artisan MCP:

No access to the live VM Service: cannot read Magic.find().rxState or any controller state.
No way to drive the UI: gestures, snapshots, screenshots, modal waits all need the Dusk extension surface.
No HTTP / log / exception inspection: Telescope's runtime feed stays invisible.
No connected REPL: every state inspection becomes a print() round-trip + hot-reload.

With artisan MCP

The artisan MCP server bridges the agent to the running app over the VM Service:

10 substrate command tools (artisan_start, artisan_stop, artisan_status, artisan_logs, artisan_restart, artisan_reload, artisan_hot_restart, artisan_doctor, artisan_list, artisan_tinker).
Plugin tools surface automatically once the consumer's bin/artisan.dart wrapper registers the provider; see each plugin's MCP tool reference for the current catalog (fluttersdk_dusk, fluttersdk_telescope).
Tool visibility filters via .artisan/mcp.json + env vars + CLI flags (Cargo-style replace for allow, union for deny).
Zero release-build impact: every artisan integration gates on kDebugMode at the consumer's main.dart.

Add use artisan to your prompt to nudge the agent toward the MCP tool surface:

Show me the current MonitorController state. use artisan

Installation

The canonical .mcp.json entry shape (written by dart run fluttersdk_artisan mcp:install):

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Every client below uses this exact shape (config-file path differs; the command + args payload does not). The cwd field must point at the project root (the directory that contains pubspec.yaml).

Cursor

Go to: Settings, Cursor Settings, MCP, Add new global MCP server.

Paste into ~/.cursor/mcp.json:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

For project-scoped use, write the same snippet to .cursor/mcp.json in the project root. The spawned process is equivalent to running dart run fluttersdk_artisan:mcp from that directory.

Claude Code

Run this command from the project root to add the artisan MCP server:

claude mcp add fluttersdk -- dart run fluttersdk_artisan:mcp

For project-scoped configuration:

claude mcp add --scope project fluttersdk -- dart run fluttersdk_artisan:mcp

You can also write .mcp.json directly with the canonical entry shape from the Installation section above.

Note: Claude Code does NOT auto-reconnect on .mcp.json edits. Run /mcp reconnect fluttersdk inside Claude Code after every edit to load the new entry.

Claude Desktop

On Windows and macOS, install the official Claude Desktop application.

Open File, Settings, Developer, MCP Servers, Edit Config to open claude_desktop_config.json:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the artisan MCP server configuration. Replace the cwd value with the absolute path to your Flutter project root, since Claude Desktop has no project context:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "/absolute/path/to/your/flutter/project"
    }
  }
}

Important: Fully quit Claude Desktop via File, Exit (closing the window just minimizes it). Restart it for the new entry to take effect.

VS Code (GitHub Copilot)

Create or edit .vscode/mcp.json in your project:

{
  "servers": {
    "fluttersdk": {
      "type": "stdio",
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

For user-level configuration, add the same entry to your VS Code settings under chat.mcp.servers.

Windsurf

Add to your Windsurf MCP configuration:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Reload the Cascade panel after editing the config; Windsurf spawns dart run fluttersdk_artisan:mcp on next request.

JetBrains IDEs (Junie / AI Assistant)

For Junie: Open Junie, click the three dots in the top right corner, then Settings, MCP Settings:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

For AI Assistant: Go to Settings, Tools, AI Assistant, MCP and add the same entry via the "as JSON" option.

Cline / Roo-Code

Add to your Cline MCP server configuration (typically ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json on macOS, equivalent paths on other OSes):

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Roo-Code reads the same mcpServers shape; point its config file at the snippet above.

OpenCode

Add to your OpenCode configuration file (~/.config/opencode/opencode.json or per-project opencode.json):

{
  "mcp": {
    "fluttersdk": {
      "type": "local",
      "command": ["dart", "run", "fluttersdk_artisan:mcp"],
      "enabled": true
    }
  }
}

The local transport spawns the Dart process directly; OpenCode inherits the working directory from where it was launched.

Gemini CLI

Add to ~/.gemini/settings.json:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Restart gemini after editing; the CLI re-spawns dart run fluttersdk_artisan:mcp on the next session.

Antigravity

Antigravity is Google's AI-powered IDE built on VS Code.

Open the Agent side panel (Cmd/Ctrl + L or View, Open View..., Agent).
Click Additional options (...) in the upper right of the Agent panel.
Select MCP Servers.
Click Manage MCP Servers.
Click View raw config in the upper right of the Manage MCPs editor view.
Add this configuration:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Tip: Install the Dart and Flutter extensions for the best development experience.

Firebase Studio

Firebase Studio is an agentic cloud-based development environment by Google.

In your Firebase Studio project, create .idx/mcp.json if it does not exist.
Add the artisan MCP configuration:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Rebuild your workspace:
- Open the Command Palette (Shift + Ctrl + P).
- Enter Firebase Studio: Rebuild Environment.

Other Clients

For any other MCP-compatible client that supports stdio transport, spawn the server with:

dart run fluttersdk_artisan:mcp

The working directory must contain a pubspec.yaml that depends on fluttersdk_artisan (directly or via a path / git dep). Clients that require a JSON entry shape accept the canonical form:

{
  "mcpServers": {
    "fluttersdk": {
      "command": "dart",
      "args": ["run", "fluttersdk_artisan:mcp"],
      "cwd": "."
    }
  }
}

Remote HTTP transport is NOT supported in V1; the artisan MCP server is stdio-only because it bridges to the local VM Service URI recorded in ~/.artisan/state.json by dart run artisan start.

Add a Rule

To make the agent reach for artisan MCP without an explicit use artisan prompt, add a rule to your AI client:

Cursor: Settings, Rules.

Claude Code: Add to CLAUDE.md in the project root.

Windsurf: Memories panel, add a workspace rule.

JetBrains AI Assistant: Settings, Tools, AI Assistant, Custom Instructions.

Example rule:

Always use artisan MCP when working with this Flutter project. Inspect running app state via artisan_tinker, drive the UI via dusk_* tools, and read HTTP / log / exception feeds via telescope_*. Prefer artisan_status before assuming the app is running.

Automated Install

The artisan CLI ships a built-in command that writes the canonical .mcp.json entry for you:

dart run fluttersdk_artisan mcp:install

The command is idempotent: pre-existing mcpServers keys are preserved, and running it twice replaces the fluttersdk key in-place without creating duplicates. By default it writes to .mcp.json in the current working directory; override the path via --path:

dart run fluttersdk_artisan mcp:install --path .vscode/mcp.json

To remove the entry later:

dart run fluttersdk_artisan mcp:uninstall

Both commands work without a running Flutter app; they only edit the JSON file.

Available Tools

The artisan MCP server surfaces 10 substrate tools plus plugin-contributed tools from any registered sibling plugin (fluttersdk_dusk, fluttersdk_telescope, or custom plugins). See tool-reference for the per-tool input schema, output shape, and example invocations of the substrate tools, plus links to each plugin's MCP tool reference for the current plugin tool catalog.

Tool visibility is filtered in three layers (file .artisan/mcp.json lowest priority, env vars middle, CLI flags highest priority; deny always wins over allow). To temporarily hide a tool surface without uninstalling the server, edit .artisan/mcp.json:

{
  "packages": {
    "allow": null,
    "deny": []
  },
  "tools": {
    "allow": null,
    "deny": []
  }
}

After editing .artisan/mcp.json or env vars, run /mcp reconnect fluttersdk inside Claude Code (or the equivalent reconnect action in your client); the server does NOT auto-reload in V1.