Grafana UI MCP Server▌

by grafana
AI-ready access to Grafana UI: full React component library—TypeScript source, MDX docs, Storybook examples, tests, and
Provides AI assistants with comprehensive access to Grafana's React component library, including TypeScript source code, MDX documentation, Storybook examples, test files, and design system tokens for building Grafana-compatible interfaces.
best for
- / Frontend developers building Grafana plugins
- / UI developers creating Grafana-compatible interfaces
- / Teams extending Grafana's design system
capabilities
- / Retrieve Grafana UI component source code
- / Access MDX documentation and usage guidelines
- / Get Storybook examples and interactive stories
- / Extract design tokens for colors, typography, and spacing
- / Search components by name and documentation content
- / Analyze component dependencies and relationships
what it does
Provides AI assistants with access to Grafana's React component library, including TypeScript source code, documentation, and design system tokens. Enables building Grafana-compatible interfaces through comprehensive component data retrieval.
about
Grafana UI MCP Server is an official MCP server published by grafana that provides AI assistants with tools and capabilities via the Model Context Protocol. AI-ready access to Grafana UI: full React component library—TypeScript source, MDX docs, Storybook examples, tests, and It is categorized under developer tools, design.
how to install
You can install Grafana UI MCP Server in your AI client of choice. Use the install panel on this page to get one-click setup for Cursor, Claude Desktop, VS Code, and other MCP-compatible clients. This server runs locally on your machine via the stdio transport.
license
MIT
Grafana UI MCP Server is released under the MIT license. This is a permissive open-source license, meaning you can freely use, modify, and distribute the software.
readme
Grafana UI MCP Server
A Model Context Protocol (MCP) server that provides AI assistants with comprehensive access to Grafana UI components, documentation, stories, and design system tokens. This server enables AI tools like Claude Desktop, Continue.dev, and other MCP-compatible clients to retrieve and work with Grafana's React component library seamlessly.
🚀 Key Features
- Complete Component Access: Get the latest Grafana UI component TypeScript source code
- Rich Documentation: Access comprehensive MDX documentation with usage guidelines
- Interactive Stories: Retrieve Storybook stories with interactive examples and controls
- Test Files: Access test files showing real usage patterns and edge cases
- Design System Integration: Get Grafana's design tokens (colors, typography, spacing, shadows)
- Dependency Analysis: Understand component relationships and dependency trees
- Advanced Search: Search components by name and documentation content
- GitHub API Integration: Efficient caching and intelligent rate limit handling
🛠️ Unified Tool Interface
This MCP server provides a single unified tool called grafana_ui that consolidates all functionality through action-based routing. This reduces complexity and makes it easier for AI agents to understand and use.
🎯 The grafana_ui Tool
All operations are performed through one tool with an action parameter:
{
"tool": "grafana_ui",
"arguments": {
"action": "get_component",
"componentName": "Button"
}
}
📋 Available Actions (11 Total)
Core Component Actions:
get_component- Get TypeScript source code for any Grafana UI componentget_demo- Get Storybook demo files showing component usagelist_components- List all available Grafana UI componentsget_metadata- Get component props, exports, and metadataget_directory- Browse the Grafana UI repository structure
Advanced Grafana Actions:
get_documentation- Get rich MDX documentation with sections and examplesget_stories- Get parsed Storybook stories with interactive controlsget_tests- Get test files showing usage patterns and edge casessearch- Search components by name and optionally by documentation contentget_theme_tokens- Get Grafana design system tokens (colors, typography, spacing, etc.)get_dependencies- Get component dependency tree analysis (shallow or deep)
✨ Benefits of the Unified Tool
- Simplified Integration: Only one tool to configure in MCP clients
- Easier for AI Agents: Reduced cognitive load with single entry point
- Better Context Management: All functionality accessible through one interface
- Parameter Validation: Comprehensive validation based on action type
- Future-Proof: Easy to add new actions without breaking changes
🔄 Migration from Previous Versions
Breaking Change: Version 2.0+ uses a unified tool interface. If you were using individual tools like
get_component,list_components, etc., you now need to use thegrafana_uitool with anactionparameter.
Before (v1.x):
{ "tool": "get_component", "arguments": { "componentName": "Button" } }
After (v2.0+):
{ "tool": "grafana_ui", "arguments": { "action": "get_component", "componentName": "Button" } }
All functionality remains the same - only the interface has changed.
📦 Quick Start
⚡ Using npx (Recommended)
The fastest way to get started - no installation required!
# Basic usage (rate limited to 60 requests/hour)
npx @shelldandy/grafana-ui-mcp-server
# With GitHub token for better rate limits (5000 requests/hour)
npx @shelldandy/grafana-ui-mcp-server --github-api-key ghp_your_token_here
# Short form
npx @shelldandy/grafana-ui-mcp-server -g ghp_your_token_here
# Using environment variable (either option works)
export GITHUB_PERSONAL_ACCESS_TOKEN=ghp_your_token_here
npx @shelldandy/grafana-ui-mcp-server
# Or using the common GITHUB_TOKEN variable
export GITHUB_TOKEN=ghp_your_token_here
npx @shelldandy/grafana-ui-mcp-server
🎯 Try it now: Run npx @shelldandy/grafana-ui-mcp-server --help to see all options!
🔧 Command Line Options
grafana-ui-mcp [options]
Options:
--github-api-key, -g <token> GitHub Personal Access Token
--grafana-repo-path, -l <path> Path to local Grafana repository (takes precedence)
--help, -h Show help message
--version, -v Show version information
Environment Variables:
GITHUB_PERSONAL_ACCESS_TOKEN Alternative way to provide GitHub token
GITHUB_TOKEN Alternative way to provide GitHub token
GRAFANA_REPO_PATH Path to local Grafana repository
Examples:
npx @shelldandy/grafana-ui-mcp-server --help
npx @shelldandy/grafana-ui-mcp-server --version
npx @shelldandy/grafana-ui-mcp-server -g ghp_1234567890abcdef
npx @shelldandy/grafana-ui-mcp-server --grafana-repo-path /path/to/grafana
npx @shelldandy/grafana-ui-mcp-server -l /path/to/grafana
GITHUB_PERSONAL_ACCESS_TOKEN=ghp_token npx @shelldandy/grafana-ui-mcp-server
GITHUB_TOKEN=ghp_token npx @shelldandy/grafana-ui-mcp-server
GRAFANA_REPO_PATH=/path/to/grafana npx @shelldandy/grafana-ui-mcp-server
🔑 GitHub API Token Setup
Why do you need a token?
- Without token: Limited to 60 API requests per hour
- With token: Up to 5,000 requests per hour
- Better reliability and faster responses
- Access to the complete Grafana UI component library
📝 Getting Your Token (2 minutes)
-
Go to GitHub Settings:
-
Generate New Token:
- Click "Generate new token (classic)"
- Add a note: "Grafana UI MCP server"
- Expiration: Choose your preference (90 days recommended)
- Scopes: ✅
public_repo(for optimal access to Grafana repository)
-
Copy Your Token:
- Copy the generated token (starts with
ghp_) - ⚠️ Save it securely - you won't see it again!
- Copy the generated token (starts with
🚀 Using Your Token
Method 1: Command Line (Quick testing)
npx @shelldandy/grafana-ui-mcp-server --github-api-key ghp_your_token_here
Method 2: Environment Variable (Recommended)
# Add to your shell profile (~/.bashrc, ~/.zshrc, etc.)
export GITHUB_PERSONAL_ACCESS_TOKEN=ghp_your_token_here
# Then simply run:
npx @shelldandy/grafana-ui-mcp-server
Method 3: Claude Desktop Configuration
{
"mcpServers": {
"grafana-ui": {
"command": "npx",
"args": ["@shelldandy/grafana-ui-mcp-server"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
}
}
}
}
✅ Verify Your Setup
# Test without token (should show rate limit warning)
npx @shelldandy/grafana-ui-mcp-server --help
# Test with token (should show success message)
npx @shelldandy/grafana-ui-mcp-server --github-api-key ghp_your_token --help
# Check your current rate limit
curl -H "Authorization: token ghp_your_token" https://api.github.com/rate_limit
🏠 Local Development Support
NEW: Work with a local Grafana repository for faster development and access to uncommitted changes!
🎯 Why Use Local Repository?
- ⚡ Faster Access: Direct filesystem reads, no network latency
- 🚫 No Rate Limits: Unlimited component access
- 🔄 Real-time Updates: See your local changes immediately
- 📡 Offline Support: Works without internet connection
- 🧪 Development Workflow: Test with modified/uncommitted components
🔧 Setup with Local Repository
-
Clone the Grafana Repository:
git clone https://github.com/grafana/grafana.git cd grafana -
Use Local Path (takes precedence over GitHub API):
# Command line option npx @shelldandy/grafana-ui-mcp-server --grafana-repo-path /path/to/grafana npx @shelldandy/grafana-ui-mcp-server -l /path/to/grafana # Environment variable export GRAFANA_REPO_PATH=/path/to/grafana npx @shelldandy/grafana-ui-mcp-server -
Claude Desktop Configuration:
{ "mcpServers": { "grafana-ui": { "command": "npx", "args": ["@shelldandy/grafana-ui-mcp-server"], "env": { "GRAFANA_REPO_PATH": "/path/to/your/grafana/repository" } } } }
🔄 Configuration Priority
The server checks sources in this order:
- Local Repository (
--grafana-repo-pathorGRAFANA_REPO_PATH) - GitHub API with Token (
--github-api-keyorGITHUB_*_TOKEN) - GitHub API without Token (rate limited to 60 requests/hour)
🛡️ Graceful Fallback
- If local file doesn't exist → Falls back to GitHub API automatically
- If local repository is invalid → Falls back to GitHub API with warning
- Source is indicated in tool responses (
"source": "local"vs"source": "github")
✅ Verify Local Setup
# Test local repository access
npx @shelldandy/grafana-ui-mcp-server --grafana-repo-path /path/to/grafana --help
# Should show: "Local Grafana repository configured: /path/to/grafana"
🛠️ Tool Usage Examples
The MCP server provides the unified grafana_ui tool for AI assistants:
Basic Component Access
// Get Button component source code
{
"tool": "grafana_ui",
"arguments": {
"action": "get_component",
"componentName": "Button"
}
}
// List all available components
{
"tool": "grafana_
---