DocuForge MCP Server
Enables AI agents to create, edit, and export IEEE research papers as PDFs, with citation management and cloud hosting capabilities via MCP.
README
DocuForge MCP Server
AI-powered document creation, editing, and PDF generation via MCP (Model Context Protocol). DocuForge allows AI agents, IDEs, and user-facing dashboards to programmatically orchestrate the creation of complex documents (specifically tailored for IEEE Research Papers), compile them to PDFs via Puppeteer, and host them.
๐ Features Added in this Project
DocuForge has been heavily upgraded with production-ready features tailored for academic and professional document workflows:
- IEEE Research Paper Pipeline: Full support for formatting documents into the official IEEE template (two-column, justified, specialized Abstract/Index Terms, and Reference sections).
- Citation Management: Native support for BibTeX importing/exporting, and inline citation marker resolution (e.g.
[cite:jones2023]auto-converts to[1]with an aggregated References list). - SQLite Persistence: Transitioned from purely in-memory storage to a production-ready SQLite database for documents, citations, and versions.
- Cloudinary Asset Hosting: Automatically uploads generated PDFs to Cloudinary when API keys are provided, returning public URLs instead of local file paths.
- Remote MCP via HTTP/SSE: Run the MCP server remotely behind an Express REST API with token authentication, allowing secure remote integration with clients like Claude Desktop.
- Multi-stage Docker Builds: Optimized
Dockerfileanddocker-compose.ymlfor running the REST API and MCP services side-by-side in production. - VS Code Extension: A functional VS Code extension bundled into the repository for IDE integration.
๐ค Quick Start for Claude Desktop (Local MCP)
Want to give Claude (or Cursor) the ability to autonomously format IEEE papers, manage citations, and export PDFs? You can run DocuForge directly via NPM!
1. Connect to Claude Desktop
Open your Claude Desktop configuration file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - Mac:
~/Library/Application Support/Claude/claude_desktop_config.json
Add the docuforge tool to your mcpServers list and add your own Gemini API key!
{
"mcpServers": {
"docuforge": {
"command": "npx",
"args": [
"-y",
"@docuforge-mcp/mcp-server"
],
"env": {
"STORAGE_TYPE": "sqlite",
"GEMINI_API_KEY": "your_gemini_api_key_here"
}
}
}
}
2. Start Creating!
Restart Claude Desktop. You will see a plug icon indicating the tools are loaded! You can now type:
"Create a new IEEE format research paper about Neural Networks, add a few sections, and export it to a PDF."
๐ Local Development & Dashboard
If you want to run the full dashboard and REST API locally:
# Run the local REST API
pnpm dev:api
# In a new terminal, run the React frontend dashboard
cd packages/dashboard
pnpm dev
โ๏ธ Cloud Deployment (Render)
This repository includes a render.yaml Blueprint file, making it extremely easy to deploy to Render.
- Push this code to your GitHub repository.
- Sign in to Render and click New+ -> Blueprint.
- Connect your GitHub repository.
- Render will automatically detect the
render.yamlfile and provision the web services. - Set your secret environment variables (
MCP_API_KEY,JWT_SECRET,CLOUDINARY_*) in the Render dashboard for the service.
๐ฆ NPM Packages
DocuForge is published on NPM and split into several modular packages:
@docuforge-mcp/core: Document/citation models, SQLite storage layer, and IEEE rendering engine.@docuforge-mcp/pdf-engine: Puppeteer-based HTML-to-PDF compilation.@docuforge-mcp/ai-integration: Gemini LLM wrapper for autonomous document improvements.@docuforge-mcp/mcp-server: Registers 9+ core MCP tools (e.g.create_ieee_paper,import_bibtex,export_pdf).
๐ System Architecture
The overarching monorepo also includes web-facing functionality:
@docuforge-mcp/rest-api: Express app serving the REST endpoints and the MCP SSE/mcpproxy.@docuforge-mcp/dashboard: Vite + React frontend.docuforge-vscode: VS Code extension client.
๐งช Testing
The repository features comprehensive testing via Vitest, encompassing unit tests, citation integration tests, and HTTP endpoint verification.
pnpm test
๐ 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.