Docfork
STDIOHTTP-SSE通过MCP提供最新文档和代码示例
通过MCP提供最新文档和代码示例

Docfork MCP pulls @latest documentation and code examples straight from the source - and adds them right into your context.
Just tell Cursor to use docfork:
Create a basic Next.js app with the App Router. use docfork
To install Docfork MCP Server for any client automatically via Smithery:
npx -y @smithery/cli@latest install @docfork/mcp --client <CLIENT_NAME> --key <YOUR_SMITHERY_KEY>
You can find your Smithery key in the Smithery.ai webpage.
Go to: Settings -> Cursor Settings -> Tools & Integrations -> Add a custom MCP server
Pasting the following config into your Cursor ~/.cursor/mcp.json file is the recommended approach. You can also install in a specific project by creating .cursor/mcp.json in your project folder. See Cursor MCP docs for more info.
{ "mcpServers": { "docfork": { "url": "https://mcp.docfork.com/mcp" } } }
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
{ "mcpServers": { "docfork": { "command": "bunx", "args": ["-y", "docfork"] } } }
{ "mcpServers": { "docfork": { "command": "deno", "args": ["run", "--allow-env", "--allow-net", "npm:docfork"] } } }
Run this command. See Claude Code MCP docs for more info.
claude mcp add --transport sse docfork https://mcp.docfork.com/sse
claude mcp add docfork -- npx -y docfork
Add this to your Claude Desktop claude_desktop_config.json file. See Claude Desktop MCP docs for more info.
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
Add this to your Windsurf MCP config. See Windsurf MCP docs for more info.
{ "mcpServers": { "docfork": { "serverUrl": "https://mcp.docfork.com/sse" } } }
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
Add this to your VS Code MCP config. See VS Code MCP docs for more info.
{ "mcpServers": { "docfork": { "type": "http", "url": "https://mcp.docfork.com/mcp" } } }
{ "servers": { "docfork": { "type": "stdio", "command": "npx", "args": ["-y", "docfork"] } } }
One-click install: → Get the Docfork Extension
Or Manual config (for power users):
{ "context_servers": { "docfork": { "command": { "path": "npx", "args": ["-y", "docfork"] }, "settings": {} } } }
Open the "Settings" page of the app, navigate to "Plugins," and enter the following JSON:
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
More info is available on BoltAI's Documentation site. For BoltAI on iOS, see this guide.
If you prefer to run the MCP server in a Docker container:
Build the Docker Image:
First, create a Dockerfile in the project root (or anywhere you prefer):
FROM node:18-alpine WORKDIR /app # Install the latest version globally RUN npm install -g docfork # Expose default port if needed (optional, depends on MCP client interaction) # EXPOSE 3000 # Default command to run the server CMD ["docfork"]
Then, build the image using a tag (e.g., docfork-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 docfork .
Configure Your MCP Client:
Update your MCP client's configuration to use the Docker command.
Example for a cline_mcp_settings.json:
{ "mcpServers": { "docfork": { "autoApprove": [], "disabled": false, "timeout": 60, "command": "docker", "args": ["run", "-i", "--rm", "docfork-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.
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/docfork/mcp": { "command": "cmd", "args": ["/c", "npx", "-y", "docfork@latest"], "disabled": false, "autoApprove": [] } } }
To configure Docfork MCP in Augment Code, follow these steps:
mcpServers array in the augment.advanced object"augment.advanced": { "mcpServers": [ { "name": "docfork", "command": "npx", "args": ["-y", "docfork"] } ] }
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.
Add this to your Roo Code MCP configuration file. See Roo Code MCP docs for more info.
{ "mcpServers": { "docfork": { "type": "streamable-http", "url": "https://mcp.docfork.com/mcp" } } }
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
Use the Add manually feature and fill in the JSON configuration information for that MCP server. For more details, visit the Trae documentation.
{ "mcpServers": { "docfork": { "url": "https://mcp.docfork.com/mcp" } } }
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
You can configure Docfork 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):
{ "mcp": { "servers": { "docfork": { "type": "http", "url": "https://mcp.docfork.com/mcp" } } } }
Or, for a local server:
{ "mcp": { "servers": { "docfork": { "type": "stdio", "command": "npx", "args": ["-y", "docfork"] } } } }
For more information and troubleshooting, refer to the Visual Studio MCP Servers documentation.
See Gemini CLI Configuration for details.
~/.gemini/settings.json (where ~ is your home directory).mcpServers object in your settings.json file:{ "mcpServers": { "docfork": { "httpUrl": "https://mcp.docfork.com/mcp" } } }
Or, for a local server:
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
If the mcpServers object does not exist, create it.
Add this to your Crush configuration file. See Crush MCP docs for more info.
{ "$schema": "https://charm.land/crush.json", "mcp": { "docfork": { "type": "http", "url": "https://mcp.docfork.com/mcp" } } }
{ "$schema": "https://charm.land/crush.json", "mcp": { "docfork": { "type": "sse", "url": "https://mcp.docfork.com/sse" } } }
{ "$schema": "https://charm.land/crush.json", "mcp": { "docfork": { "type": "stdio", "command": "npx", "args": ["-y", "docfork"] } } }
You can easily install Docfork through the Cline MCP Server Marketplace by following these instructions:
To configure Docfork MCP in Zencoder, follow these steps:
{ "command": "npx", "args": ["-y", "docfork@latest"] }
Once the MCP server is added, you can easily continue using it.
Add this to your Amazon Q Developer CLI configuration file. See Amazon Q Developer CLI docs for more details.
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork@latest"] } } }
See Qodo Gen docs for more details.
{ "mcpServers": { "docfork": { "url": "https://mcp.docfork.com/mcp" } } }
See JetBrains AI Assistant Documentation for more details.
Settings -> Tools -> AI Assistant -> Model Context Protocol (MCP)+ Add.Command in the top-left corner of the dialog and select the As JSON option from the listOK{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
Apply to save changes.Settings -> Tools -> Junie -> MCP SettingsSee Warp Model Context Protocol Documentation for details.
Settings > AI > Manage MCP servers.+ Add button.{ "Docfork": { "command": "npx", "args": ["-y", "docfork"], "env": {}, "working_directory": null, "start_on_launch": true } }
Save to apply the changes.Add this to your Opencode configuration file. See Opencode MCP docs docs for more info.
"mcp": { "docfork": { "type": "remote", "url": "https://mcp.docfork.com/mcp", "enabled": true } }
{ "mcp": { "docfork": { "type": "local", "command": ["npx", "-y", "docfork"], "enabled": true } } }
Add the following configuration to the mcp section of your Copilot Coding Agent configuration file Repository->Settings->Copilot->Coding agent->MCP configuration:
{ "mcpServers": { "docfork": { "type": "http", "url": "https://mcp.docfork.com/mcp", "tools": ["get-library-docs"] } } }
For more information, see the official GitHub documentation.
See Kiro Model Context Protocol Documentation for details.
Kiro > MCP Servers+ Add button.{ "mcpServers": { "Docfork": { "command": "npx", "args": ["-y", "docfork"], "env": {}, "disabled": false, "autoApprove": [] } } }
Save to apply the changes.See OpenAI Codex for more information.
Add the following configuration to your OpenAI Codex MCP server settings:
[mcp_servers.docfork] args = ["-y", "docfork"] command = "npx"
See LM Studio MCP Support for more information.
Program (right side) > Install > Edit mcp.json.{ "mcpServers": { "Docfork": { "command": "npx", "args": ["-y", "docfork"] } } }
Save to apply the changes.Program, or by clicking the plug icon at the bottom of the chat box.See Local and Remote MCPs for Perplexity for more information.
Perplexity > SettingsConnectors.Add Connector.Advanced.Docfork{ "args": ["-y", "docfork"], "command": "npx", "env": {} }
Save.Docfork MCP provides different tools depending on the client type:
search-docs: Search for documentation across the web, GitHub, and other sources.
query (required): Query for documentation. Include language/framework/library names.tokens (optional): Token budget control for response size.read-docs: Read the content of a documentation URL as markdown/text.
url (required): The URL of the webpage to read (typically from search-docs results).For OpenAI ChatGPT integration, Docfork provides OpenAI MCP specification-compliant tools:
search: Search for documents using semantic search. Returns a list of relevant search results.
query (required): Search query string. Natural language queries work best for semantic search.fetch: Retrieve complete document content by ID for detailed analysis and citation.
id (required): URL or unique identifier for the document to fetch.Note:
The OpenAI tools (searchandfetch) automatically format their responses for ChatGPT connectors and are compatible with deep research workflows.
If you don't want to add use docfork to every prompt, you can define a simple rule from your Cursor Settings > Rules section in Cursor (or the equivalent in your MCP client) to auto-invoke Docfork on any code question:
--- alwaysApply: true --- when the user requests code examples, setup or configuration steps, or library/API documentation use docfork.
From then on you'll get Docfork's docs in any related conversation without typing anything extra. You can add your use cases to the match part.
When you know exactly which library you want to use, be specific in your prompts. This helps Docfork find the right documentation faster and more accurately:
implement basic authentication with supabase. use docfork
create a Next.js middleware for rate limiting. use docfork
configure Tailwind CSS with custom typography. use docfork
The more specific you are about the library and what you want to accomplish, the better documentation you'll receive.
Clone the project and install dependencies:
npm i
Build:
npm run build
The Docfork MCP server supports the following environment variables:
DEFAULT_MINIMUM_TOKENS: Set the minimum token count for documentation retrieval (default: 10000)The following environment variables are only relevant when running the server as an HTTP/SSE service (not for standard npx usage):
MCP_TRANSPORT: Set the transport type for MCP communication (default: stdio, options: streamable-http, stdio, sse)PORT: Set the port number for HTTP/SSE transport (default: 3000, only used when MCP_TRANSPORT is streamable-http or sse)Standard node server configuration (most common):
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "docfork@latest"], "env": { "DEFAULT_MINIMUM_TOKENS": "10000" } } } }
HTTP/SSE server configuration (for custom deployments):
These environment variables are used when you're running your own instance of the Docfork server, not when connecting to remote servers. For remote server connections, use the URL-based configurations shown earlier in this README (e.g., "url": "https://mcp.docfork.com/mcp").
If you're self-hosting and want to run the server with HTTP/SSE transport:
# Set environment variables and run MCP_TRANSPORT=streamable-http PORT=3000 npx -y docfork@latest
{ "mcpServers": { "docfork": { "command": "node", "args": ["/path/to/folder/docfork/dist/index.js"] } } }
npm run inspect
If you encounter ERR_MODULE_NOT_FOUND, try using bunx instead of npx:
{ "mcpServers": { "docfork": { "command": "bunx", "args": ["-y", "docfork"] } } }
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": { "docfork": { "command": "npx", "args": ["-y", "--node-options=--experimental-vm-modules", "docfork"] } } }
Use the --experimental-fetch flag to bypass TLS-related problems:
{ "mcpServers": { "docfork": { "command": "npx", "args": ["-y", "--node-options=--experimental-fetch", "docfork"] } } }
@latest to the package namebunx as an alternative to npxdeno as another alternativeDocfork is an open, community-driven catalogue. Although we review submissions, we make no warranties—express or implied—about the accuracy, completeness, or security of any linked documentation or code. Projects listed here are created and maintained by their respective authors, not by Docfork.
If you spot content that is suspicious, inappropriate, or potentially harmful, please contact us.
By using Docfork, you agree to do so at your own discretion and risk.
Stay in the loop and meet the community:
MIT