Skip to main content
This quickstart guide will get you up and running with Context7 MCP server in your preferred AI coding assistant.
API Key Recommended: Get a free API key at context7.com/dashboard for higher rate limits. Context7 works without an API key, but you’ll have lower rate limits.
The fastest way to set up Context7 is using the ctx7 CLI:
1

Run setup command

npx ctx7 setup
This command will:
  • Authenticate via OAuth
  • Generate an API key automatically
  • Configure the MCP server for your detected agents
  • Add a rule to auto-invoke Context7
2

Select your AI coding assistant

The CLI will detect installed agents (Cursor, Claude Code, OpenCode) and let you choose which to configure.You can also target a specific agent:
npx ctx7 setup --cursor
npx ctx7 setup --claude
npx ctx7 setup --opencode
3

Start using Context7

Restart your AI coding assistant and add use context7 to your prompts:
Create a Next.js middleware that checks for a valid JWT in cookies. use context7
Or set up a rule to auto-invoke Context7 for all code-related questions.

​
Manual Setup

If you prefer to configure manually, follow the instructions for your AI coding assistant.
For project-specific configuration instead of global, use the --project flag:
npx ctx7 setup --project
Go to: Settings → Cursor Settings → MCP → Add new global MCP serverAdd this configuration to your ~/.cursor/mcp.json file:
{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "CONTEXT7_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

​
Local Server

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}
Since Cursor 1.0, you can also use the one-click install button in the README.
claude mcp add --scope user --header "CONTEXT7_API_KEY: YOUR_API_KEY" --transport http context7 https://mcp.context7.com/mcp

​
Local Server

claude mcp add --scope user context7 -- npx -y @upstash/context7-mcp --api-key YOUR_API_KEY
Remove --scope user to install for the current project only.
Add this to your OpenCode configuration file:
"mcp": {
  "context7": {
    "type": "remote",
    "url": "https://mcp.context7.com/mcp",
    "headers": {
      "CONTEXT7_API_KEY": "YOUR_API_KEY"
    },
    "enabled": true
  }
}

​
Local Server

{
  "mcp": {
    "context7": {
      "type": "local",
      "command": ["npx", "-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"],
      "enabled": true
    }
  }
}

​
Add an Auto-Invoke Rule

To avoid typing use context7 in every prompt, add a rule to your AI coding assistant: Cursor: Cursor Settings → Rules Claude Code: Create or edit CLAUDE.md in your project OpenCode: Configuration rules section

​
Example Rule

Always use Context7 MCP when I need library/API documentation, code generation, 
setup or configuration steps without me having to explicitly ask.

​
Using Library IDs

If you already know the exact library you want, specify its Context7 ID to skip the library-matching step: 
Implement basic authentication with Supabase. use library /supabase/supabase for API and docs.
The slash syntax tells the MCP tool exactly which library to load docs for.

​
Specifying Versions

To get documentation for a specific library version, mention it in your prompt: 
How do I set up Next.js 14 middleware? use context7
Context7 will automatically match the appropriate version.

​
Verify Installation

1

Restart your AI coding assistant

Close and reopen Cursor, Claude Code, or OpenCode to load the new MCP configuration.
2

Test with a simple prompt

Show me how to create a basic Express.js server. use context7
You should see Context7 being invoked to fetch Express.js documentation.
3

Check MCP server logs

Most AI coding assistants show MCP server status in their settings or debug panel. Look for Context7 in the list of active servers.

​
Troubleshooting

Common Issues:
  • Make sure your API key starts with ctx7sk
  • For local servers, ensure Node.js 18+ is installed
  • Check that your AI coding assistant supports MCP (most recent versions do)
  • Verify the configuration file path is correct for your OS
If Context7 isn’t working:
  1. Check API Key: Get a fresh key from context7.com/dashboard
  2. Check Configuration: Ensure JSON syntax is valid (no trailing commas)
  3. Check Logs: Look for error messages in your AI assistant’s MCP logs
  4. Try Remote Server: Remote servers are more reliable than local ones

​
Next Steps

Installation Guide

Learn about SDK, CLI, and AI SDK tools installation

API Reference

Explore the REST API for custom integrations

Build docs developers (and LLMs) love

Get started for freeTalk to us