Context7 MCP - Up-to-date Code Docs For Any Prompt

❌ Without Context7

LLMs rely on outdated or generic information about the libraries you use. You get:

• ❌ Code examples are outdated and based on year-old training data
• ❌ Hallucinated APIs that don't even exist
• ❌ Generic answers for old package versions

✅ With Context7

Context7 MCP pulls up-to-date, version-specific documentation and code examples straight from the source — and places them directly into your prompt.

Add use context7 to your prompt (or set up a rule to auto-invoke):

Create a Next.js middleware that checks for a valid JWT in cookies
and redirects unauthenticated users to `/login`. use context7

Configure a Cloudflare Worker script to cache
JSON API responses for five minutes. use context7

Context7 fetches up-to-date code examples and documentation right into your LLM's context.

• 1️⃣ Write your prompt naturally
• 2️⃣ Tell the LLM to use context7 (or set up a rule once)
• 3️⃣ Get working code answers

No tab-switching, no hallucinated APIs that don't exist, no outdated code generation.

[!NOTE] This repository hosts the source code of Context7 MCP server. The supporting components — API backend, parsing engine, and crawling engine — are private and not part of this release.

📚 Adding Projects

Check out our project addition guide to learn how to add (or update) your favorite libraries to Context7.

🛠️ Installation

Requirements

• Node.js >= v20.18.1
• Cursor, Claude Code, VSCode, Devin Desktop or another MCP Client
• Context7 API Key (Optional) for higher rate limits and private repositories (Get yours by creating an account at context7.com/dashboard)

[!TIP] Recommended Post-Setup: Add a Rule to Auto-Invoke Context7

After installing Context7 (see instructions below), enhance your workflow by adding a rule so you don't have to type use context7 in every prompt. Define a simple rule in your MCP client's rule section to automatically invoke Context7 on any code question:

• For Devin Desktop, in .devin/rules/ directory
• For Cursor, from Cursor Settings > Rules section
• For Claude Code, in CLAUDE.md file
• Or the equivalent in your MCP client

Example Rule:

Always use context7 when I need code generation, setup or configuration steps, or
library/API documentation. This means you should automatically use the Context7 MCP
tools to resolve library id and get library docs without me having to explicitly ask.

From then on, you'll get Context7's docs in any related conversation without typing anything extra. You can alter the rule to match your use cases.

To install Context7 MCP Server for any client automatically via Smithery:

npx -y @smithery/cli@latest install @upstash/context7-mcp --client <CLIENT_NAME> --key <YOUR_SMITHERY_KEY>

You can find your Smithery key in the Smithery.ai webpage.

Go to: Settings -> Cursor Settings -> MCP -> Add new global MCP server

Pasting the following configuration into your Cursor ~/.cursor/mcp.json file is the recommended approach. You may also install in a specific project by creating .cursor/mcp.json in your project folder. See Cursor MCP docs for more info.

Since Cursor 1.0, you can click the install button below for instant one-click installation.

Cursor Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Cursor Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Run this command. See Claude Code MCP docs for more info.

Claude Code Local Server Connection

claude mcp add --scope user context7 -- npx -y @upstash/context7-mcp --api-key YOUR_API_KEY

Claude Code Remote Server Connection

claude mcp add --scope user --header "Authorization: Bearer YOUR_API_KEY" --transport http context7 https://mcp.context7.com/mcp

Remove --scope user to install for the current project only.

Run this command in your terminal. See Amp MCP docs for more info.

Without API Key (Basic Usage)

amp mcp add context7 https://mcp.context7.com/mcp

With API Key (Higher Rate Limits & Private Repos)

amp mcp add context7 --header "Authorization=Bearer YOUR_API_KEY" https://mcp.context7.com/mcp

Add this to your Devin Desktop MCP config file. See Devin Desktop MCP docs for more info.

Devin Desktop Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "serverUrl": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Devin Desktop Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Add this to your VS Code MCP config file. See VS Code MCP docs for more info.

VS Code Remote Server Connection

"mcp": {
  "servers": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

VS Code Local Server Connection

"mcp": {
  "servers": {
    "context7": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

You can easily install Context7 through the Cline MCP Server Marketplace by following these instructions:

1. Open Cline.
2. Click the hamburger menu icon (☰) to enter the MCP Servers section.
3. Use the search bar within the Marketplace tab to find Context7.
4. Click the Install button.

Or you can directly edit MCP servers configuration:

1. Open Cline.
2. Click the hamburger menu icon (☰) to enter the MCP Servers section.
3. Choose Remote Servers tab.
4. Click the Edit Configuration button.
5. Add context7 to mcpServers:

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp",
      "type": "streamableHttp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

It can be installed via Zed Extensions or you can add this to your Zed settings.json. See Zed Context Server docs for more info.

{
  "context_servers": {
    "Context7": {
      "source": "custom",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

To configure Context7 MCP in Augment Code, you can use either the graphical interface or manual configuration.

A. Using the Augment Code UI

1. Click the hamburger menu.

2. Select Settings.

3. Navigate to the Tools section.

4. Click the + Add MCP button.

5. Enter the following command:

   npx -y @upstash/context7-mcp@latest

6. Name the MCP: Context7.

7. Click the Add button.

Once the MCP server is added, you can start using Context7's up-to-date code documentation features directly within Augment Code.

B. Manual Configuration

1. Press Cmd/Ctrl Shift P or go to the hamburger menu in the Augment panel
2. Select Edit Settings
3. Under Advanced, click Edit in settings.json
4. Add the server configuration to the mcpServers array in the augment.advanced object

"augment.advanced": {
  "mcpServers": [
    {
      "name": "context7",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  ]
}

Once the MCP server is added, restart your editor. If you receive any errors, check the syntax to make sure closing brackets or commas are not missing.

You can configure the Context7 MCP server in Kilo Code using either the UI or by editing your project's MCP configuration file.

Kilo Code supports two configuration levels:

• Global MCP Configuration — stored in mcp_settings.json
• Project-level MCP Configuration — stored in .kilocode/mcp.json (recommended)

If a server is defined in both places, the project-level configuration overrides the global one.

Configure via Kilo Code UI

1. Open Kilo Code.
2. Click the Settings icon in the top-right corner.
3. Navigate to Settings → MCP Servers.
4. Click Add Server.
5. Choose HTTP Server (Streamable HTTP Transport).
6. Enter the details:

URL https://mcp.context7.com/mcp

Headers → Add Header

• Key: Authorization
• Value: Bearer YOUR_API_KEY

7. Click Save.
8. Ensure the server toggle is enabled.
9. If needed, click Refresh MCP Servers to reload the configuration.

Manual Configuration (.kilocode/mcp.json)

To configure the server at the project level (recommended for team environments), create the following file:

.kilocode/mcp.json:

{
  "mcpServers": {
    "context7": {
      "type": "streamable-http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      },
      "alwaysAllow": [],
      "disabled": false
    }
  }
}

Replace YOUR_API_KEY with your actual Context7 API key.

After saving the file:

• Open Settings → MCP Servers

• Click Refresh MCP Servers

Kilo Code will automatically detect and load the configuration.

Add this to your Antigravity MCP config file. See Antigravity MCP docs for more info.

Google Antigravity Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "serverUrl": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Google Antigravity Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Add this to your Roo Code MCP configuration file. See Roo Code MCP docs for more info.

Roo Code Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "type": "streamable-http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Roo Code Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

See Gemini CLI Configuration for details.

1. Open the Gemini CLI settings file. The location is ~/.gemini/settings.json (where ~ is your home directory).
2. Add the following to the mcpServers object in your settings.json file:

{
  "mcpServers": {
    "context7": {
      "httpUrl": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY",
        "Accept": "application/json, text/event-stream"
      }
    }
  }
}

Or, for a local server:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

If the mcpServers object does not exist, create it.

See Qwen Coder MCP Configuration for details.

1. Open the Qwen Coder settings file. The location is ~/.qwen/settings.json (where ~ is your home directory).
2. Add the following to the mcpServers object in your settings.json file:

{
  "mcpServers": {
    "context7": {
      "httpUrl": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY",
        "Accept": "application/json, text/event-stream"
      }
    }
  }
}

Or, for a local server:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

If the mcpServers object does not exist, create it.

Remote Server Connection

Open Claude Desktop and navigate to Settings > Connectors > Add Custom Connector. Enter the name as Context7 and the remote MCP server URL as https://mcp.context7.com/mcp.

Local Server Connection

Open Claude Desktop developer settings and edit your claude_desktop_config.json file to add the following configuration. See Claude Desktop MCP docs for more info.

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Add this to your Opencode configuration file. See Opencode MCP docs for more info.

Opencode Remote Server Connection

"mcp": {
  "context7": {
    "type": "remote",
    "url": "https://mcp.context7.com/mcp",
    "headers": {
      "Authorization": "Bearer YOUR_API_KEY"
    },
    "enabled": true
  }
}

Opencode Local Server Connection

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

See OpenAI Codex for more information.

Add the following configuration to your OpenAI Codex MCP server settings:

Local Server Connection

[mcp_servers.context7]
args = ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
command = "npx"
startup_timeout_ms = 20_000

Remote Server Connection

[mcp_servers.context7]
url = "https://mcp.context7.com/mcp"
http_headers = { "Authorization" = "Bearer YOUR_API_KEY" }

Optional troubleshooting — only if you see startup "request timed out" or "not found program". Most users can ignore this.

• First try: increase startup_timeout_ms to 40_000 and retry.
• Windows quick fix (absolute npx path + explicit env):

[mcp_servers.context7]
command = "C:\\Users\\yourname\\AppData\\Roaming\\npm\\npx.cmd"
args = [
  "-y",
  "@upstash/context7-mcp",
  "--api-key",
  "YOUR_API_KEY"
]
env = { SystemRoot="C:\\Windows", APPDATA="C:\\Users\\yourname\\AppData\\Roaming" }
startup_timeout_ms = 40_000

• macOS quick fix (use Node + installed package entry point):

[mcp_servers.context7]
command = "/Users/yourname/.nvm/versions/node/v22.14.0/bin/node"
args = ["/Users/yourname/.nvm/versions/node/v22.14.0/lib/node_modules/@upstash/context7-mcp/dist/index.js",
  "--transport",
  "stdio",
  "--api-key",
  "YOUR_API_KEY"
]

Notes: Replace yourname with your OS username. Explicitly setting APPDATA and SystemRoot is essential because these are required by npx on Windows but not set by certain versions of OpenAI Codex mcp clients by default.

See JetBrains AI Assistant Documentation for more details.

1. In JetBrains IDEs, go to Settings -> Tools -> AI Assistant -> Model Context Protocol (MCP)
2. Click + Add.
3. Click on Command in the top-left corner of the dialog and select the As JSON option from the list
4. Add this configuration and click OK

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

5. Click Apply to save changes.
6. The same way context7 could be added for JetBrains Junie in Settings -> Tools -> Junie -> MCP Settings

See Kiro Model Context Protocol Documentation for details.

1. Navigate Kiro > MCP Servers
2. Add a new MCP server by clicking the + Add button.
3. Paste one of the configurations below:

Kiro Remote Server Connection

{
  "mcpServers": {
    "Context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

To use an API key in Kiro, add:

"headers": {
  "Authorization": "Bearer YOUR_API_KEY"
}

You can create an API key at context7.com/dashboard for authenticated usage and higher rate limits.

Kiro Local Server Connection

{
  "mcpServers": {
    "Context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"],
      "env": {},
      "disabled": false,
      "autoApprove": []
    }
  }
}

4. Click Save to apply the changes.

Use the Add manually feature and fill in the JSON configuration information for that MCP server. For more details, visit the Trae documentation.

Trae Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

Trae Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Use these alternatives to run the local Context7 MCP server with other runtimes. These examples work for any client that supports launching a local MCP server via command + args.

Bun

{
  "mcpServers": {
    "context7": {
      "command": "bunx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Deno

{
  "mcpServers": {
    "context7": {
      "command": "deno",
      "args": [
        "run",
        "--allow-env=NO_DEPRECATION,TRACE_DEPRECATION",
        "--allow-net",
        "npm:@upstash/context7-mcp"
      ]
    }
  }
}

If you prefer to run the MCP server in a Docker container:

1. Build the Docker Image:

   First, create a Dockerfile in the project root (or anywhere you prefer):

   FROM node:20-alpine
   
   WORKDIR /app
   
   # Install the latest version globally
   RUN npm install -g @upstash/context7-mcp
   
   # Expose default port if needed (optional, depends on MCP client interaction)
   # EXPOSE 3000
   
   # Default command to run the server
   CMD ["context7-mcp", "--transport", "stdio"]

   Then, build the image using a tag (e.g., context7-mcp). Make sure Docker Desktop (or the Docker daemon) is running. Run the following command in the same directory where you saved the Dockerfile:

   docker build -t context7-mcp .

2. Configure Your MCP Client:

   Update your MCP client's configuration to use the Docker command.

   Example for a cline_mcp_settings.json:

   {
     "mcpServers": {
       "Сontext7": {
         "autoApprove": [],
         "disabled": false,
         "timeout": 60,
         "command": "docker",
         "args": ["run", "-i", "--rm", "context7-mcp"],
         "transportType": "stdio"
       }
     }
   }

   Note: This is an example configuration. Please refer to the specific examples for your MCP client (like Cursor, VS Code, etc.) earlier in this README to adapt the structure (e.g., mcpServers vs servers). Also, ensure the image name in args matches the tag used during the docker build command.

   If you use the Docker MCP Toolkit image (mcp/context7) with a stdio-based client, set MCP_TRANSPORT=stdio so the container starts with stdio transport instead of its HTTP default. For Cline, Roo Code, and Claude Desktop, use:

   {
     "mcpServers": {
       "context7": {
         "command": "docker",
         "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT=stdio", "mcp/context7"]
       }
     }
   }

   For VS Code, use the same Docker command in the servers format:

   {
     "servers": {
       "context7": {
         "type": "stdio",
         "command": "docker",
         "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT=stdio", "mcp/context7"]
       }
     }
   }

   Keep using the remote server URL for HTTP-based clients.

Install the context7.mcpb file under the mcpb folder and add it to your client. For more information, please check out MCP bundles docs.

The configuration on Windows is slightly different compared to Linux or macOS (Cline is used in the example). The same principle applies to other editors; refer to the configuration of command and args.

{
  "mcpServers": {
    "github.com/upstash/context7-mcp": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"],
      "disabled": false,
      "autoApprove": []
    }
  }
}

Add this to your Amazon Q Developer CLI configuration file. See Amazon Q Developer CLI docs for more details.

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

See Warp Model Context Protocol Documentation for details.

1. Navigate Settings > AI > Manage MCP servers.
2. Add a new MCP server by clicking the + Add button.
3. Paste the configuration given below:

{
  "Context7": {
    "command": "npx",
    "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"],
    "env": {},
    "working_directory": null,
    "start_on_launch": true
  }
}

4. Click Save to apply the changes.

Using Context7 with Copilot Coding Agent

Add the following configuration to the mcp section of your Copilot Coding Agent configuration file Repository->Settings->Copilot->Coding agent->MCP configuration:

{
  "mcpServers": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      },
      "tools": ["get-library-docs", "resolve-library-id"]
    }
  }
}

For more information, see the official GitHub documentation.

1. Open the Copilot CLI MCP config file. The location is ~/.copilot/mcp-config.json (where ~ is your home directory).
2. Add the following to the mcpServers object in your mcp-config.json file:

{
  "mcpServers": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      },
      "tools": ["get-library-docs", "resolve-library-id"]
    }
  }
}

Or, for a local server:

{
  "mcpServers": {
    "context7": {
      "type": "local",
      "command": "npx",
      "tools": ["get-library-docs", "resolve-library-id"],
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

If the mcp-config.json file does not exist, create it.

See LM Studio MCP Support for more information.

One-click install:

Manual set-up:

1. Navigate to Program (right side) > Install > Edit mcp.json.
2. Paste the configuration given below:

{
  "mcpServers": {
    "Context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

3. Click Save to apply the changes.
4. Toggle the MCP server on/off from the right hand side, under Program, or by clicking the plug icon at the bottom of the chat box.

You can configure Context7 MCP in Visual Studio 2022 by following the Visual Studio MCP Servers documentation.

Add this to your Visual Studio MCP config file (see the Visual Studio docs for details):

{
  "inputs": [],
  "servers": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Or, for a local server:

{
  "mcp": {
    "servers": {
      "context7": {
        "type": "stdio",
        "command": "npx",
        "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
      }
    }
  }
}

For more information and troubleshooting, refer to the Visual Studio MCP Servers documentation.

Add this to your Crush configuration file. See Crush MCP docs for more info.

Crush Remote Server Connection (HTTP)

{
  "$schema": "https://charm.land/crush.json",
  "mcp": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Crush Local Server Connection

{
  "$schema": "https://charm.land/crush.json",
  "mcp": {
    "context7": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Open the "Settings" page of the app, navigate to "Plugins," and enter the following JSON:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Once saved, enter in the chat get-library-docs followed by your Context7 documentation ID (e.g., get-library-docs /nuxt/ui). More information is available on BoltAI's Documentation site. For BoltAI on iOS, see this guide.

Edit your Rovo Dev CLI MCP config by running the command below -

acli rovodev mcp

Example config -

Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

To configure Context7 MCP in Zencoder, follow these steps:

1. Go to the Zencoder menu (...)
2. From the dropdown menu, select Agent tools
3. Click on the Add custom MCP
4. Add the name and server configuration from below, and make sure to hit the Install button

{
  "command": "npx",
  "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
}

Once the MCP server is added, you can easily continue using it.

See Qodo Gen docs for more details.

1. Open Qodo Gen chat panel in VSCode or IntelliJ.
2. Click Connect more tools.
3. Click + Add new MCP.
4. Add the following configuration:

Qodo Gen Local Server Connection

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Qodo Gen Remote Server Connection

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

See Local and Remote MCPs for Perplexity for more information.

1. Navigate Perplexity > Settings
2. Select Connectors.
3. Click Add Connector.
4. Select Advanced.
5. Enter Server Name: Context7
6. Paste the following JSON in the text area:

{
  "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"],
  "command": "npx",
  "env": {}
}

7. Click Save.

Factory's droid supports MCP servers through its CLI. See Factory MCP docs for more info.

Factory Remote Server Connection (HTTP)

Run this command in your terminal:

droid mcp add context7 https://mcp.context7.com/mcp --type http --header "Authorization: Bearer YOUR_API_KEY"

Or without an API key (basic usage with rate limits):

droid mcp add context7 https://mcp.context7.com/mcp --type http

Factory Local Server Connection (Stdio)

Run this command in your terminal:

droid mcp add context7 "npx -y @upstash/context7-mcp" --env CONTEXT7_API_KEY=YOUR_API_KEY

Once configured, Context7 tools will be available in your droid sessions. Type /mcp within droid to manage servers, authenticate, and view available tools.

Emdash is an orchestration layer for running multiple coding agents in parallel. Provider-agnostic, worktree-isolated, and local-first. Emdash supports Context7 MCP to enable Context7 for your agents.

What Emdash provides:

• Global toggle: Settings → MCP → "Enable Context7 MCP"
• Per-workspace enable: The Context7 button in the ProviderBar (off by default). First click enables it for that workspace. Clicking again disables it.
• ProviderBar: The Context7 button shows status, a short explanation, and a link to docs

What you still need to do: Configure your coding agent (Codex, Claude Code, Cursor, etc.) to connect to Context7 MCP. Emdash does not modify your agent's config. See the respective MCP configuration sections above for your agent (e.g., OpenAI Codex, Claude Code, Cursor).

See the Emdash repository for more information.

Use the Autohand Code CLI to add the local Context7 server:

autohand mcp add context7 npx -y @upstash/context7-mcp --api-key YOUR_API_KEY

Add --scope project before context7 to save the server in the current project's .autohand configuration instead of your user configuration.

For basic usage without an API key, you can connect to the remote server instead:

autohand mcp add --transport http context7 https://mcp.context7.com/mcp

🔨 Available Tools

Context7 MCP provides the following tools that LLMs can use:

• resolve-library-id: Resolves a general library name into a Context7-compatible library ID.

  • libraryName (required): The name of the library to search for

• get-library-docs: Fetches documentation for a library using a Context7-compatible library ID.

  • context7CompatibleLibraryID (required): Exact Context7-compatible library ID (e.g., /mongodb/docs, /vercel/next.js)
  • topic (optional): Focus the docs on a specific topic (e.g., "routing", "hooks")
  • page (optional, default 1): Page number for pagination (1-10). If the context is not sufficient, try page=2, page=3, etc. with the same topic.

🛟 Tips

Add a Rule

To avoid typing use context7 in every prompt, you can add a rule to your MCP client that automatically invokes Context7 for code-related questions. See the recommended setup in the Installation section for detailed instructions and example rules.

Use Library Id

If you already know exactly which library you want to use, add its Context7 ID to your prompt. That way, Context7 MCP server can skip the library-matching step and directly continue with retrieving docs.

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.

HTTPS Proxy

If you are behind an HTTP proxy, Context7 uses the standard https_proxy / HTTPS_PROXY environment variables.

💻 Development

Clone the project and install dependencies:

bun i

Build:

bun run build

Run the server:

bun run dist/index.js

CLI Arguments

context7-mcp accepts the following CLI flags:

• --transport <stdio|http> – Transport to use (stdio by default). Use http for remote HTTP server or stdio for local integration.
• --port <number> – Port to listen on when using http transport (default 3000).
• --api-key <key> – API key for authentication (or set CONTEXT7_API_KEY env var). You can get your API key by creating an account at context7.com/dashboard.

Example with HTTP transport and port 8080:

bun run dist/index.js --transport http --port 8080

Another example with stdio transport:

bun run dist/index.js --transport stdio --api-key YOUR_API_KEY

Environment Variables

You can use the CONTEXT7_API_KEY environment variable instead of passing the --api-key flag. This is useful for:

• Storing API keys securely in .env files
• Integration with MCP server setups that use dotenv
• Tools that prefer environment variable configuration

Note: The --api-key CLI flag takes precedence over the environment variable when both are provided.

Example with .env file:

# .env
CONTEXT7_API_KEY=your_api_key_here

Example MCP configuration using environment variable:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp"],
      "env": {
        "CONTEXT7_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

OpenTelemetry observability

Context7 instruments individual MCP requests and notifications at the SDK transport boundary, including messages inside a valid batch and MCP v2 subscriptions/listen operations handled by the SDK entry layer. Requests rejected by the SDK's HTTP envelope and protocol-version validation before dispatch remain visible in normal HTTP/gateway telemetry, but are not reported as MCP operations. Observed operations follow the development-status OpenTelemetry MCP semantic conventions for server metrics and spans. Trace context is extracted from the traceparent, tracestate, and baggage fields in MCP params._meta as defined by SEP-414.

The HTTP transport exposes metrics in Prometheus format on a dedicated listener at 127.0.0.1:9464/metrics by default. The production Docker image explicitly binds that listener to 0.0.0.0 so an internal Prometheus pod scraper or PodMonitor can reach it. The stdio transport does not open a telemetry port. Keeping this listener separate from the public MCP port prevents the metrics endpoint from being routed through a catch-all gateway rule. On SIGTERM, SIGINT, or SIGHUP, both transports use a bounded shutdown path that stops serving, closes active MCP connections and subscriptions, and best-effort flushes externally installed SDK metric and trace providers before exit. Stdio EOF triggers the same path.

The exporter uses the standard OpenTelemetry Prometheus settings:

• OTEL_EXPORTER_PROMETHEUS_HOST changes the bind address (default 127.0.0.1; the Docker image sets 0.0.0.0).
• OTEL_EXPORTER_PROMETHEUS_PORT changes the port (default 9464).
• OTEL_METRICS_EXPORTER=none or OTEL_SDK_DISABLED=true disables the embedded exporter.

OTEL_SDK_DISABLED=true is the hard-off switch: provider modules are not loaded and MCP transports and handlers are not wrapped, preserving the baseline request path. In contrast, OTEL_METRICS_EXPORTER=none disables only the embedded Prometheus bootstrap, so a provider installed by a Node preload can still receive the MCP signals.

Exporter bind or configuration failures are logged but do not prevent the MCP endpoint from starting. If a Node preload has already registered global OpenTelemetry providers, they take precedence. The embedded Prometheus listener is not started when a global MeterProvider exists, and MCP spans are exported through the preload's TracerProvider. This supports an OpenTelemetry Node SDK or Kubernetes auto-instrumentation without creating a second provider in the application. When an external SDK owns the provider, configure its Node runtime instrumentation there as well; the application does not register a duplicate collector.

It reports bounded-cardinality counters, histograms, and in-flight gauges for MCP methods, subscriptions, tool outcomes, authentication outcomes, Context7 upstream requests, and Node runtime saturation. Prometheus receives these metric families:

• mcp_server_operation_duration (its _count series is the MCP operation count, and tool-call series include the context7_mcp_tool_outcome label)
• mcp_server_session_duration for real stateful stdio sessions (stateless HTTP request transports are intentionally excluded)
• context7_mcp_operations_active
• context7_mcp_subscriptions_active and context7_mcp_subscription_duration
• context7_mcp_upstream_requests_total and context7_mcp_upstream_request_duration
• context7_mcp_authentication_attempts_total and context7_mcp_authentication_duration
• context7_mcp_upstream_requests_active and context7_mcp_authentication_active
• nodejs_eventloop_*, v8js_gc_duration, v8js_memory_heap_*, and v8js_resource_active from the official OpenTelemetry Node runtime instrumentation

Tool outcomes on the standard MCP operation metric distinguish success, not_found, and error. An acknowledged subscriptions/listen operation is timed through its acknowledgement; the separate subscription metrics track the active stream and its bounded terminal outcome. Upstream outcomes distinguish HTTP, response-decoding, network, timeout, and cancellation failures and include both the bounded status-code class and the exact numeric HTTP status. Authentication reports accepted, missing, invalid, and unexpected-error outcomes. The OAuth authorization-server metadata proxy caps its upstream fetch at 10 seconds and returns 502 if that dependency times out.

The labels intentionally exclude API keys, client IPs, queries, library IDs, session IDs, and raw error text. Expose port 9464 only to your Prometheus scraper or ServiceMonitor, not through the public MCP ingress.

Signal ownership with an Envoy gateway

Do not treat mcp_server_operation_duration_count as another HTTP request counter. An Envoy Gateway observes HTTP envelopes, while this metric observes JSON-RPC requests and notifications after SDK dispatch. A valid batch is one HTTP request but several MCP operations, and HTTP requests rejected before MCP dispatch never increment the MCP metric.

Context7 deliberately does not register generic inbound HTTP server metrics. Keep the following signals in the existing Envoy scrape instead of collecting them again from the application:

• downstream HTTP request/response totals, status classes, duration, active requests, connections, resets, and gateway timeouts (envoy_http_*_downstream_*)
• Envoy-to-MCP backend request totals, status codes, duration, active/pending requests, connection failures, retries, resets, timeouts, and circuit-breaker overflows (envoy_cluster_upstream_*)
• Envoy process health and resource metrics

The application exporter owns only signals the ingress gateway cannot provide: MCP method and protocol semantics (including batches, notifications, and active MCP v2 subscriptions), tool and authentication outcomes, MCP-to-Context7 API calls, and Node event-loop/V8 health. In the Kubernetes deployment Envoy is a Gateway API proxy rather than a sidecar in the MCP pod, so context7_mcp_upstream_* describes the MCP server's outbound Context7 API dependency, not Envoy's inbound MCP backend cluster. Pod and container CPU, memory, network, and restart metrics should continue to come from the Kubernetes monitoring stack.

See the Envoy HTTP connection manager statistics and upstream cluster statistics for the proxy-owned metric families.

For a replicated Kubernetes deployment, discover and scrape every MCP pod directly. Do not use one static, load-balanced Service target: successive scrapes can reach different replicas and produce incomplete per-process counters and runtime series. For an annotation-based kubernetes-pods scrape job, add the following fields to the MCP workload's pod template:

spec:
  template:
    metadata:
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/port: "9464"
        prometheus.io/path: /metrics
    spec:
      containers:
        - name: mcp
          ports:
            - name: metrics
              containerPort: 9464
              protocol: TCP

Prometheus will then scrape http://<mcp-pod-ip>:9464/metrics for each replica. Declaring EXPOSE 9464 in the image does not add the Kubernetes containerPort metadata. The scrape interval is controlled by Prometheus; the exporter does not impose one. If the monitoring stack uses the Prometheus Operator instead, configure the equivalent per-pod endpoint with a PodMonitor.

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["tsx", "/path/to/folder/context7/src/index.ts", "--api-key", "YOUR_API_KEY"]
    }
  }
}

OAuth Authentication

Context7 MCP server supports OAuth 2.0 authentication for MCP clients that implement the MCP OAuth specification.

To use OAuth, change the endpoint from /mcp to /mcp/oauth in your client configuration:

- "url": "https://mcp.context7.com/mcp"
+ "url": "https://mcp.context7.com/mcp/oauth"

Note: OAuth is not supported with stdio transport. For local MCP connections, use API key authentication instead.

npx -y @modelcontextprotocol/inspector npx @upstash/context7-mcp

🚨 Troubleshooting

If you encounter ERR_MODULE_NOT_FOUND, try using bunx instead of npx:

{
  "mcpServers": {
    "context7": {
      "command": "bunx",
      "args": ["-y", "@upstash/context7-mcp"]
    }
  }
}

This often resolves module resolution issues in environments where npx doesn't properly install or resolve packages.

For errors like Error: Cannot find module 'uriTemplate.js', try the --experimental-vm-modules flag:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "--node-options=--experimental-vm-modules", "@upstash/[email protected]"]
    }
  }
}

Use the --experimental-fetch flag to bypass TLS-related problems:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "--node-options=--experimental-fetch", "@upstash/context7-mcp"]
    }
  }
}

1. Try adding @latest to the package name
2. Use bunx as an alternative to npx
3. Consider using deno as another alternative
4. Ensure you're using Node.js v20 or higher for native fetch support

⚠️ Disclaimer

1- Context7 projects are community-contributed and while we strive to maintain high quality, we cannot guarantee the accuracy, completeness, or security of all library documentation. Projects listed in Context7 are developed and maintained by their respective owners, not by Context7. If you encounter any suspicious, inappropriate, or potentially harmful content, please use the "Report" button on the project page to notify us immediately. We take all reports seriously and will review flagged content promptly to maintain the integrity and safety of our platform. By using Context7, you acknowledge that you do so at your own discretion and risk.

2- This repository hosts the MCP server’s source code. The supporting components — API backend, parsing engine, and crawling engine — are private and not part of this release.

🤝 Connect with Us

Stay updated and join our community:

• 📢 Follow us on X for the latest news and updates
• 🌐 Visit our Website
• 💬 Join our Discord Community

📺 Context7 In Media

• Better Stack: "Free Tool Makes Cursor 10x Smarter"
• Cole Medin: "This is Hands Down the BEST MCP Server for AI Coding Assistants"
• Income Stream Surfers: "Context7 + SequentialThinking MCPs: Is This AGI?"
• Julian Goldie SEO: "Context7: New MCP AI Agent Update"
• JeredBlu: "Context 7 MCP: Get Documentation Instantly + VS Code Setup"
• Income Stream Surfers: "Context7: The New MCP Server That Will CHANGE AI Coding"
• AICodeKing: "Context7 + Cline & RooCode: This MCP Server Makes CLINE 100X MORE EFFECTIVE!"
• Sean Kochel: "5 MCP Servers For Vibe Coding Glory (Just Plug-In & Go)"

Vercel Marketplace OIDC

Vercel Marketplace resources can call the remote MCP server without a long-lived Context7 API key. Obtain a fresh per-resource access token from Vercel's Marketplace integration runtime and send it as the bearer credential:

import { getIntegrationToken } from "@vercel/integrations";

const authorization = `Bearer ${await getIntegrationToken("context7")}`;

The getIntegrationToken signature is based on Vercel's current provider specification and may change before Marketplace OIDC is generally available.

Use authorization as the Authorization header when creating the MCP HTTP transport. Create the transport inside the request that uses it so a short-lived token is not retained across function invocations. Its resource claim must match the Context7 resource created during Marketplace provisioning.

The hosted MCP service must be configured with the exact issuer and audience assigned by Vercel when the Context7 Marketplace product is created:

VERCEL_MARKETPLACE_OIDC_ISSUER=https://integrations.vercel.com/oac_...
VERCEL_MARKETPLACE_OIDC_AUDIENCE=https://integrations.vercel.com/context7/icfg_...

📄 License

MIT