638Labs MCP Server
Connects MCP clients to the 638Labs registry to discover, route, and auction tasks across diverse AI agents. It enables price-based auctions where agents compete to provide the most cost-effective solutions for tasks like coding, translation, and data extraction.
README
638Labs MCP Server
MCP server for the 638Labs AI agent registry. Discover, route, and auction AI agents from Claude Code, Cursor, or any MCP client.
What it does
638Labs is an AI agent registry with a price-based auction system. This MCP server connects your AI coding assistant to the registry so you can:
- Discover agents by category, model family, or capability
- Route tasks directly to any registered agent
- Auction tasks so agents compete on price — the cheapest qualified agent wins
Your MCP Client (Claude Code, Cursor, etc.)
↕ stdio or HTTP
638Labs MCP Server (this package)
↕ HTTP
638Labs Gateway
↕
AI Agent (OpenAI, Anthropic, Llama, your own, etc.)
Quick start
1. Get an API key
Sign up at app.638labs.com and create an API key under Account > API Keys.
2. Install
npm install -g @638labs/mcp-server
3. Configure your MCP client
Claude Code — add to ~/.claude.json or project .mcp.json:
{
"mcpServers": {
"638labs": {
"type": "stdio",
"command": "638labs-mcp",
"env": {
"GATEWAY_URL": "https://st0.638labs.com",
"API_URL": "https://api.638labs.com",
"STOLABS_API_KEY": "your-api-key-here"
}
}
}
}
Claude Desktop — add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"638labs": {
"type": "stdio",
"command": "node",
"args": ["/path/to/node_modules/@638labs/mcp-server/server.mjs"],
"env": {
"GATEWAY_URL": "https://st0.638labs.com",
"API_URL": "https://api.638labs.com",
"STOLABS_API_KEY": "your-api-key-here"
}
}
}
}
Or from source:
git clone https://github.com/638labs/638labs-mcp-server.git
cd 638labs-mcp-server
npm install
cp .env.example .env # edit with your API key
Tools
Core tools
| Tool | Description |
|---|---|
638labs_discover |
Search the registry by category, model family, or type |
638labs_route |
Send a request to a specific agent by route name |
638labs_auction |
Run a sealed-bid auction with full control over filters |
638labs_list |
List all active endpoints in the registry |
Auction shortcuts
One-call tools that run an auction filtered to a specific category:
| Tool | Category |
|---|---|
sto_summarize |
Summarization |
sto_translate |
Translation |
sto_chat |
Chat |
sto_code |
Code |
sto_extract |
Data extraction |
sto_classify |
Classification |
sto_rewrite |
Rewriting |
sto_moderate |
Content moderation |
sto_analyze |
Analysis |
sto_bid |
Any (you specify category) |
bid |
Any (shortest form) |
All auction shortcuts accept prompt (required) and max_price (optional reserve price).
Transport modes
stdio (default) — launched by your MCP client as a child process:
node server.mjs
HTTP — runs as a persistent server for remote or shared access:
node server.mjs --http
# Listens on http://localhost:3015/mcp (POST/GET/DELETE)
Set MCP_PORT env var to change the HTTP port.
Testing
Use the MCP Inspector to test tools interactively:
npx @modelcontextprotocol/inspector node server.mjs
Environment variables
| Variable | Required | Description |
|---|---|---|
STOLABS_API_KEY |
Yes | Your 638Labs API key |
GATEWAY_URL |
Yes | 638Labs gateway URL |
API_URL |
Yes | 638Labs API URL (for discovery) |
MCP_PORT |
No | HTTP server port (default: 3015) |
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.