com.docfork/docfork-mcp MCP Server
com.docfork/docfork-mcp
DEPRECATED and shut down — was a docs server for AI agents; use io.github.docfork/docfork instead.
What is the com.docfork/docfork-mcp MCP server?
Docfork MCP is a deprecated server that provided up-to-date, hallucination-free documentation for AI coding agents like Cursor, Claude Code, and Windsurf by exposing `search_docs` and `fetch_doc` tools. As of June 14, 2026, Docfork has shut down entirely — the MCP server, CLI, API keys, and hosted endpoints (including https://mcp.docfork.com) are all permanently offline. Users are directed to the successor package io.github.docfork/docfork instead.
Docfork MCP was designed to solve the problem of AI agents hallucinating APIs or writing code against stale documentation by serving current, version-accurate docs directly inside IDEs (Cursor, Claude Code, Windsurf, and 29+ supported clients). It offered semantic + BM25 hybrid search over a curated catalog of framework docs, plus the ability to index custom public/private GitHub repos as searchable libraries. This specific package (com.docfork/docfork-mcp) is now deprecated and non-functional — the entire Docfork service shut down on June 14, 2026, and all endpoints, API keys, and CLI setup commands are permanently offline. Anyone looking for this functionality should use the replacement package io.github.docfork/docfork instead.
How to install com.docfork/docfork-mcp
Copy-paste configuration for popular MCP clients.
{
"mcpServers": {
"docfork-mcp": {
"command": "npx",
"args": [
"-y",
"docfork"
]
}
}
}{
"mcpServers": {
"docfork-mcp": {
"command": "npx",
"args": [
"-y",
"docfork"
]
}
}
}{
"mcpServers": {
"docfork-mcp": {
"command": "npx",
"args": [
"-y",
"docfork"
]
}
}
}{
"servers": {
"docfork-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"docfork"
]
}
}
}claude mcp add docfork-mcp -- npx -y docforkTools & capabilities
Tools this server exposes to the agent.
search_docs— Returns ranked documentation sections with titles, URLs, and relevance scores for a given library and query.fetch_doc— Returns full rendered markdown content from a documentation URL found via search_docs.
com.docfork/docfork-mcp MCP server FAQ
No. As of June 14, 2026, Docfork has fully shut down — the MCP server, CLI, and all hosted endpoints (including https://mcp.docfork.com) are permanently offline, and API keys no longer work.
The README states this package is deprecated in favor of io.github.docfork/docfork, though note the underlying Docfork service itself has also shut down.
It served up-to-date documentation to AI coding agents (Cursor, Claude Code, Windsurf, and others) via two tools, search_docs and fetch_doc, to prevent hallucinated or outdated API usage.
Historically it offered a free tier of 1,000 requests/month per organization, with paid plans for teams to share API keys and Cabinets, but this is no longer relevant since the service is shut down.
Previously it supported sign-in on first use with no API key required for individual use, remote HTTP connections via https://mcp.docfork.com/mcp, and OAuth via the /mcp/oauth endpoint — none of which function anymore.
According to the README, code and prompts never left the user's machine; only queries and library names were sent to Docfork's servers, and queries were not stored.
README (reference)
Source of truth, from the repository.
<p align="center">Up-to-date docs for AI coding agents.</p>[!CAUTION] Docfork has shut down as of June 14, 2026. The MCP server, CLI, and all hosted endpoints are offline and will not return. API keys no longer work, and
npx dgrep setupwill fail. Thank you to everyone who used the service.
AI agents hallucinate APIs, bloat context with stale docs, and write code against outdated signatures. Docfork serves up-to-date documentation directly in Cursor, Claude Code, and Windsurf.
Without Docfork
app.use('/api/*', jwt({ secret: ... }))
- ^^^ removed in Hono v4
With Docfork
app.use('/api/*', bearerAuth({ verifyToken: ... }))
+ ^^^ current API, Hono v4.2
Get Started
npx dgrep setup
Installs the Docfork MCP server in your IDE. Detects your installed agents and writes the config file; sign in to Docfork on first use, no API key needed. Target one with --agent claude-code (also: cursor, codex, opencode, vscode, windsurf, amp, factory, zed).
Your agent now has two tools:
| Tool | Returns |
|---|---|
search_docs | Ranked documentation sections with titles, URLs, and relevance scores. |
fetch_doc | Full rendered markdown content from a documentation URL. |
No prompt suffix needed:
Set up server-side rendering with Next.js App Router.
Or search from the terminal:
dgrep search "middleware redirect based on authentication" -l vercel/next.js
dgrep search "server actions with forms" -l vercel/next.js
Quickstart → · dgrep docs → · CLI reference →
Your own docs
Index any public or private GitHub repository as a custom library. Your internal APIs, SDKs, and runbooks become searchable by your agents — same pipeline as public libraries. GitHub integration setup →
Teams
Free: 1,000 requests/month per organization. For team rollout, commit the MCP config to your repo:
// .cursor/mcp.json (committed to git, picked up by every engineer)
{
"mcpServers": {
"docfork": {
"url": "https://mcp.docfork.com/mcp",
"headers": {
"DOCFORK_API_KEY": "YOUR_TEAM_API_KEY"
}
}
}
}
Share API keys and Cabinets across your organization. Docfork doesn't store your code or prompts. Security → · Pricing →
MCP Setup
[!TIP] Run
npx dgrep setupto install automatically (use--agent claude-codeto target one). Manual config below for other clients.
Cursor — <a href="https://cursor.com/en/install-mcp?name=docfork&config=eyJ1cmwiOiJodHRwczovL21jcC5kb2Nmb3JrLmNvbS9tY3AifQ%3D%3D"><img src="https://cursor.com/deeplink/mcp-install-dark.svg" height="20" alt="Add to Cursor"/></a>
{
"mcpServers": {
"docfork": {
"url": "https://mcp.docfork.com/mcp",
"headers": {
"DOCFORK_API_KEY": "YOUR_API_KEY"
}
}
}
}
Claude Code
claude mcp add --transport http docfork https://mcp.docfork.com/mcp/oauth
OpenCode
{
"mcp": {
"docfork": {
"type": "remote",
"url": "https://mcp.docfork.com/mcp",
"headers": { "DOCFORK_API_KEY": "YOUR_API_KEY" },
"enabled": true,
},
},
}
Don't see your client? Setup guides for all 29 supported clients →
OAuth Authentication
Docfork supports MCP OAuth specs. Change your endpoint to use OAuth:
- "url": "https://mcp.docfork.com/mcp"
+ "url": "https://mcp.docfork.com/mcp/oauth"
Note: OAuth is for remote HTTP connections only. View full OAuth guide →
Agent Rule
Add a rule so your agent calls Docfork MCP automatically. Full rule and IDE-specific setup →
[!NOTE] Add Rule to Cursor (One-Click)
Claude Code — add to your CLAUDE.md:
## Docfork policy
Use Docfork MCP `search_docs` and `fetch_doc` tools for library/API docs, setup, and configuration questions.
- Start `library` with a short name or keyword (e.g., `nextjs`, `zod`). Use the `owner/repo` from the result URL for follow-up calls, never guess it upfront.
- After finding a relevant result, call `fetch_doc` to get the full content. Search results are summaries only.
- Prefer Docfork results over training data when they conflict.
<details>
<summary>Full rule (all clients)</summary>
When writing or debugging code that involves third-party libraries, frameworks, or APIs, use Docfork MCP `search_docs` and `fetch_doc` tools rather than relying on training data.
**Two defaults to follow every time:**
- Start `library` with a short name or keyword (e.g., `nextjs`, `zod`). Use the `owner/repo` from the result URL for follow-up calls, never guess it upfront.
- After finding a relevant result, call `fetch_doc` to get the full content. Search results are summaries only.
Skip Docfork when:
- Language built-ins, general algorithms, syntax stable across versions
- Code or docs the user has already provided in context
When uncertain, default to using Docfork.
</details>
FAQ
How is Docfork different from Context7? Both provide MCP servers and CLIs for searching library documentation. Here are the key differences:
- Stack scoping.
dgrep initreads yourpackage.jsonand scopes all searches to your declared dependencies. Cabinets let you version-pin those libraries across a team. - Resolve once, search many.
dgrep initresolves package names to canonical identifiers once and caches the mapping in.dgrep/config.json. No per-query resolution step. - Hybrid search. Semantic search and BM25 run in parallel, fused via Reciprocal Rank Fusion. AST-aware chunking preserves function boundaries.
Does Docfork store my code or prompts? Your code and prompts never leave your machine. At search time, only the query and library name are sent to Docfork — queries are not stored. Indexed documentation content lives in an upstream vector store; private library content is end-to-end encrypted and deleted atomically when you remove the library. Security →
What libraries are supported? Docfork maintains a curated catalog of popular frameworks. Add any public or private GitHub repository as a custom library. Add custom libraries →
Docs
Community
- Changelog
- X (Twitter)
- Found an issue? Open a GitHub issue or contact us.
Star History
License
MIT
Related MCP servers
Scrapling MCP Server
Give your AI agent stealth web scraping with Cloudflare bypass and CSS selection, powered by Scrapling.
Chrome DevTools MCP
Give your AI coding agent full control of a live Chrome browser for automation, debugging, and performance analysis.
com.puter/mcp-server
Let AI agents manage your Puter files, websites, and serverless workers over MCP.
Browser automation for AI agents via MCP, powering ByteDance's Agent TARS hybrid GUI/DOM browser control.
Run arbitrary shell commands from an MCP-connected AI agent.
Filesystem access MCP server from ByteDance's UI-TARS/Agent TARS ecosystem.