vecr.io
Start now

Get MCP for Any Data Beta

Provide a GitHub repo, website URL, or text data and get an MCP server with search tools. Use it in Cursor, Claude Desktop, and any MCP-compatible app.

  • Transform any data source into an MCP server
  • Works with Cursor, Claude Desktop, and MCP-compatible tools
  • From data to MCP server in minutes
API Docs Community

Get MCP Server for Any Data Source

Provide your GitHub repository, website URL, or text documents and we'll create an MCP server with search tools. Connect it to your favorite MCP-compatible applications.

For Developers

Connect your GitHub repositories to MCP. Search your codebase, documentation, and issues directly from Cursor, Claude Desktop, or any MCP app.

Code search without leaving your IDE

For Product Teams

Turn your product docs, help center, or website into an MCP server. Team members can access knowledge directly in their development tools.

Instant knowledge access in dev tools

For Business Users

Transform your documentation, knowledge base, or content into searchable MCP tools. No technical setup required - just provide the data source.

No technical skills needed

MCP Server Setup in Minutes

Connect your MCP server to any compatible application. Use the search tools to access your data directly from Cursor, Claude Desktop, or any MCP-enabled environment.

MCP Configuration

1. Add to your MCP settings 

# claude_desktop_config.json
{
  "mcpServers": {
    "vecr": {
      "command": "npx",
      "args": ["@modelcontextprotocol/server-fetch"],
      "env": {
        "VECR_SERVER_URL": "wss://mcp.vecr.io/YOUR_INTEGRATION_ID",
        "VECR_API_TOKEN": "your_api_token"
      }
    }
  }
}

2. Available tools

# Search your data
search(query: string): SearchResults

# List available resources  
list_resources(): Resource[]

# Get specific resource
get_resource(uri: string): ResourceContent

3. Use in any MCP app 

# Works in Cursor, Claude Desktop, and more
# Just ask: "Search my docs for authentication setup"
# Or: "Find examples of error handling in my codebase"

Key Features

  • MCP Protocol Support

    Native Model Context Protocol integration for seamless tool access

  • Universal Data Sources

    GitHub repos, websites, documents - turn any data into MCP tools

  • App Integration Ready

    Works with Cursor, Claude Desktop, and any MCP-compatible application

Ready to Launch?

Create your first MCP server and start using it in your favorite development environment today.

Simple, Transparent Pricing

Start for free and scale as your needs grow. No surprises, just pay for what you use.

Free to Try

Developer Plan

$0 / month

Perfect for testing and small projects. No credit card required.

  • 500 API calls per month
  • Upload up to 10 pages
  • Standard RAG capabilities
  • Community support
Get Started Free
Most Flexible

Usage-Based Plan

Custom pricing

Tailored pricing based on your specific needs and usage volume.

  • Unlimited API calls
  • Unlimited document uploads
  • Advanced RAG configurations
  • Priority support
  • Custom integrations available
Contact for Pricing

Pricing is usage-based and scales with your needs. We work with you to create a custom plan that makes sense for your specific use case. No unexpected charges.

Frequently Asked Questions

Common questions about MCP servers and vecr.io

An MCP (Model Context Protocol) server is a standardized interface that connects AI tools like Cursor and Claude Desktop to external data sources. MCP servers expose tools, resources, and prompts that AI assistants can use to search, retrieve, and interact with your data. Think of it as a universal adapter between your content and AI applications.
vecr.io generates MCP servers automatically from your data. You provide a GitHub repository URL, website URL, or upload documents, and vecr.io indexes the content using vector embeddings. It then creates an SSE-based MCP endpoint that any MCP-compatible tool can connect to. The entire process takes minutes and requires no coding.
vecr.io supports three types of data sources: GitHub repositories (source code, documentation, READMEs, issues), websites (any publicly accessible URL), and documents (markdown, text files, PDFs). You can combine multiple sources in a single integration to create a unified search layer across all your content.
MCP servers work with any application that supports the Model Context Protocol. Popular MCP-compatible tools include Cursor (AI code editor), Claude Desktop (Anthropic's desktop app), Claude Code (CLI tool), Windsurf, and many other AI development tools. vecr.io's MCP servers use standard SSE transport, so they work with any compliant client.
A REST API exposes fixed endpoints for CRUD operations, while an MCP server exposes tools, resources, and prompts designed for AI consumption. MCP servers use a standardized protocol that AI tools understand natively, so you don't need to write integration code. The AI assistant discovers available tools automatically and uses them based on context.
Not with vecr.io. Traditional MCP server creation requires writing TypeScript or Python code, setting up vector storage, building an embedding pipeline, and deploying infrastructure. vecr.io handles all of this automatically — you just provide your data source and get a working MCP endpoint. For developers who want full control, the MCP SDK supports custom server development.
vecr.io offers three embedding models: BGE-base (768 dimensions) for general English text and code, BGE-M3 (1024 dimensions) for multilingual content and longer documents, and Cohere Embed 4 (1536 dimensions) for mixed content including text, images, tables, and code. You choose the model when creating an integration.
Yes. vecr.io's free tier includes 500 API calls per month and up to 10 pages of indexed content. This is enough to test the full workflow — create an integration, index your data, generate an MCP token, and connect from Cursor or Claude Desktop. For larger datasets, usage-based pricing scales with your needs.

Create account

Start now with a free account

Already have an account?