Vega-Lite MCP Server
The Vega-Lite MCP Server provides AI assistants with comprehensive access to Vega-Lite documentation, examples, validation tools, and schema information.
Search documentation: Search through Vega-Lite documentation for information about charts, encodings, marks, and more
Retrieve examples: Get Vega-Lite example specifications by category or type (bar, line, scatter, area, histogram, heatmap, interactive, etc.) with optional search terms
Validate specifications: Validate Vega-Lite JSON specifications to check for errors and ensure correctness
Access schema information: Get detailed information about Vega-Lite schema properties and overall structure (mark, encoding, data, transform, etc.)
Provides access to Vega-Lite documentation, examples, and validation tools, enabling AI assistants to search documentation, retrieve visualization examples by category, validate specifications, and access schema information for creating data visualizations.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Vega-Lite MCP Server show me a bar chart example with tooltips"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Vega-Lite & Deneb MCP Server
A Model Context Protocol (MCP) server that provides AI assistants with access to Vega-Lite and Deneb documentation, examples, and validation tools.
What it does
This server enables AI assistants like Claude to:
π Search Vega-Lite and Deneb documentation
π Get visualization examples by category
β Validate Vega-Lite specifications
π Access schema information
πΌ Power BI integration via Deneb documentation
Related MCP server: BI Chart MCP Server
Quick Start
1. Install
# Clone the repository
git clone https://github.com/inteligencianegociosmmx/vegaLite_mcp_server.git
cd vegaLite_mcp_server
# Install dependencies and build
npm install
npm run build2. Configure
Claude Code (VS Code) - Recommended β
# One command installation
claude mcp add vegalite node /path/to/vegaLite_mcp_server/dist/index.jsThen completely close and reopen VS Code (important - don't just reload).
Claude Desktop
Edit your config file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.jsonLinux:
~/.config/Claude/claude_desktop_config.json
Add:
{
"mcpServers": {
"vegalite": {
"command": "node",
"args": ["/absolute/path/to/vegaLite_mcp_server/dist/index.js"]
}
}
}Restart Claude Desktop.
3. Verify
Claude Code: Type /mcp - you should see "vegalite" listed
Claude Desktop: Look for the π¨ tools indicator
4. Try it
Search for "bar chart" in Vega-LiteShow me a scatter plot exampleValidate this spec: {"$schema": "https://vega.github.io/schema/vega-lite/v5.json", "mark": "bar", ...}Available Tools
Tool | What it does | Example |
| Search Vega-Lite and Deneb docs | "How do filter transforms work?" or "How does Deneb work with Power BI?" |
| Get examples by category | "Show me a line chart example" |
| Validate JSON specs | "Validate this Vega-Lite spec: {...}" |
| Get schema details | "What encoding channels are available?" |
π Deneb Support
Deneb is a certified custom visual for Microsoft Power BI that uses Vega-Lite. This server now includes Deneb documentation to help Power BI users:
Understand how to use Vega-Lite in Power BI
Learn about Power BI-specific features (cross-filtering, tooltips, etc.)
Get examples specific to the Power BI context
Try: "How do I use cross-filtering in Deneb?" or "What's the difference between Vega-Lite and Deneb?"
Troubleshooting
Server not showing up?
Check the path is absolute and correct
Test manually:
node /path/to/dist/index.js(should say "running on stdio")For Claude Code: Completely close and reopen VS Code (not just reload)
Verify with
/mcp(Claude Code) or look for tools indicator (Claude Desktop)
Tools not working?
npm run build # Rebuild the projectRemove the server:
claude mcp remove vegalite # Claude Code
# Or manually delete from config fileDevelopment
npm run watch # Auto-rebuild on changes
npm run scrape # Fetch latest Vega-Lite docsProject Structure
src/
βββ index.ts # Main MCP server
βββ tools/
β βββ search.ts # Documentation search
β βββ examples.ts # Example retrieval
β βββ validate.ts # Spec validation
βββ utils/
βββ scraper.ts # Documentation scraperManual Configuration
Claude Code (Manual)
Press
Ctrl+Shift+P(orCmd+Shift+Pon Mac)Type:
Claude Code: Edit MCP SettingsAdd:
{
"mcpServers": {
"vegalite": {
"command": "node",
"args": ["/absolute/path/to/dist/index.js"]
}
}
}Save and completely close/reopen VS Code
Other MCP Clients
Add the same configuration to your client's MCP settings file. Check your client's documentation for the config file location.
Contributing
Contributions welcome! Please submit a Pull Request.
See CONTRIBUTING.md for development guidelines.
Future Features
For information about deploying as a remote server, see docs/DEPLOYMENT.md.
License
MIT License - See LICENSE
Links
Issues: https://github.com/inteligencianegociosmmx/vegaLite_mcp_server/issues
Vega-Lite Docs: https://vega.github.io/vega-lite/
MCP Docs: https://modelcontextprotocol.io
Built with Model Context Protocol β’ Powered by Vega-Lite
Available Tools
4 toolsget_exampleB
Retrieve Vega-Lite example specifications by category or type
Inputs: category (required), search.
get_schema_infoC
Get information about Vega-Lite schema properties and structure
Inputs: property.
search_docsC
Search through Vega-Lite documentation for information about charts, encodings, marks, and more
Inputs: query (required).
validate_specC
Validate a Vega-Lite specification and check for errors
Inputs: spec (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: get_example retrieves examples, get_schema_info provides schema details, search_docs searches documentation, and validate_spec validates specifications. There is no overlap in functionality, making tool selection straightforward for an agent.
All tool names follow a consistent verb_noun pattern (get_example, get_schema_info, search_docs, validate_spec) with clear, descriptive verbs. This uniformity enhances readability and predictability across the tool set.
With 4 tools, the server is well-scoped for its purpose of supporting Vega-Lite chart development. Each tool serves a distinct, essential function (examples, schema, docs, validation), and there are no extraneous tools, making the count appropriate.
The tool set covers key aspects of Vega-Lite usage: examples, schema info, documentation search, and spec validation. A minor gap exists in lacking tools for generating or modifying specifications, but agents can work around this using the provided tools effectively.
Maintenance
Related MCP Connectors
Search VeChain documentation, query on-chain data, and fetch fee suggestions with direct links toβ¦
Astra DB tools for agents: schema, find, vector search, guarded writes, and interactive views.
Versioned documentation registry and semantic search for AI tools and coding assistants.
Exactly 50 data transformation and live web verification tools for AI agents.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides tools for AI assistants to access the Agent Web Protocol (AWP) specification, validate agent.json files, and generate protocol-compliant configurations. It enables developers to integrate the AWP standard into their websites through natural language prompts and automated validation.8 npmMIT
- AlicenseNot gradedqualityDmaintenanceEnables business intelligence chart creation and visualization using Vega-Lite, with data loading and processing capabilities.13Apache 2.0
- FlicenseAqualityDmaintenanceProvides AI agents with D3.js API documentation and example code from d3js.org and Observable gallery, enabling module discovery, documentation search, and example extraction.51-
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to list, find, fetch, and look up JSON Schema documents from the SchemaStore catalog.273 npmMIT