PDF Gen Studio MCP Server
Enables PDF and image generation from templates, JSON, HTML, or URLs through the PDF Gen Studio API. Supports rendering, template management, and multiple output formats.
README
PDF Gen Studio MCP Server
A Model Context Protocol (MCP) server for PDF Gen Studio - Generate PDFs and images from templates, JSON, HTML, or URLs.
Features
- 📄 Template Rendering - Render saved templates with dynamic data injection
- 🔧 JSON Rendering - Convert JSON design documents to PDF/images
- 🌐 HTML Rendering - Convert HTML/CSS content to PDF/images
- 📸 URL Rendering - Screenshot any website URL as PDF/images
- 📋 Template Management - List and retrieve your saved templates
Installation
Using npm (Global)
npm install -g @pdfgenstudio/mcp
Using npx (No Install)
npx @pdfgenstudio/mcp
From Source
git clone https://github.com/Majid9287/pdfgenstudio-mcp.git
cd pdfgenstudio-mcp
npm install
npm run build
Using Docker
docker pull mcp/pdfgenstudio:latest
Or build locally:
docker build -t mcp/pdfgenstudio .
Configuration
Environment Variables
Set your PDF Gen Studio API key:
export PDFGENSTUDIO_API_KEY="your-api-key-here"
Or on Windows (PowerShell):
$env:PDFGENSTUDIO_API_KEY = "your-api-key-here"
Or on Windows (Command Prompt):
set PDFGENSTUDIO_API_KEY=your-api-key-here
Optional Configuration
| Variable | Description | Default |
|---|---|---|
PDFGENSTUDIO_API_KEY |
Your PDF Gen Studio API key | Required |
PDFGENSTUDIO_BASE_URL |
API base URL | https://api.pdfgenstudio.com |
Usage
As stdio Server (Default)
# Using global install
pdfgenstudio-mcp
# Using npx
npx @pdfgenstudio/mcp
# With explicit transport
pdfgenstudio-mcp --transport stdio
As HTTP Server
# Start HTTP server on port 3100
pdfgenstudio-mcp --transport http --port 3100
# Or with custom port
pdfgenstudio-mcp -t http -p 8080
Development & Testing
# Test with fastmcp CLI
npm run test
# Inspect with MCP Inspector
npm run inspect
🔌 Client Integrations
Claude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json
Using npx (Recommended)
{
"mcpServers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Using Local Installation
{
"mcpServers": {
"pdfgenstudio": {
"command": "node",
"args": ["/absolute/path/to/pdfgenstudio-mcp/dist/index.js"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
After editing, restart Claude Desktop completely.
Docker Desktop (MCP Toolkit)
PDF Gen Studio MCP is available in Docker Desktop's MCP Toolkit catalog.
Option 1: Using Docker Desktop MCP Toolkit UI
- Open Docker Desktop
- Go to Extensions → MCP Toolkit
- Search for "pdfgenstudio" in the catalog
- Click Install and configure your API key
Option 2: Manual Configuration
Add to your Claude Desktop configuration to use the Docker image:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"pdfgenstudio": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "PDFGENSTUDIO_API_KEY=your-api-key-here",
"mcp/pdfgenstudio:latest"
]
}
}
}
Running Docker Container Directly
# Run in stdio mode (default) - for MCP client integration
docker run -it --rm \
-e PDFGENSTUDIO_API_KEY="your-api-key-here" \
mcp/pdfgenstudio:latest
# Run in HTTP mode - for local development/testing
docker run -it --rm -p 3100:3100 \
-e PDFGENSTUDIO_API_KEY="your-api-key-here" \
mcp/pdfgenstudio:latest node dist/index.js --transport http --port 3100
Using Docker Compose
# Create .env file with your API key
echo "PDFGENSTUDIO_API_KEY=your-api-key-here" > .env
# Run in stdio mode (default)
docker-compose up -d pdfgenstudio-mcp
# Run in HTTP mode (for local testing)
docker-compose --profile http up -d pdfgenstudio-mcp-http
VS Code (GitHub Copilot)
VS Code supports MCP servers through the GitHub Copilot extension.
Step 1: Enable MCP in VS Code
- Open VS Code Settings (
Ctrl+,orCmd+,) - Search for
github.copilot.chat.experimental.mcp - Enable the setting
Step 2: Configure MCP Server
Create or edit .vscode/mcp.json in your workspace:
{
"servers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Or add to your VS Code User Settings (settings.json):
{
"github.copilot.chat.experimental.mcp": true,
"mcp": {
"servers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
}
Using Local Installation in VS Code
{
"servers": {
"pdfgenstudio": {
"command": "node",
"args": ["C:/path/to/pdfgenstudio-mcp/dist/index.js"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Cursor
Cursor has built-in MCP support. Configure it in Cursor settings.
Step 1: Open Cursor Settings
- Open Cursor
- Go to
Settings→Cursor Settings→MCP - Or press
Ctrl+Shift+P/Cmd+Shift+Pand search "MCP"
Step 2: Add MCP Server
Add the following configuration:
{
"mcpServers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Alternative: Edit Config File Directly
macOS: ~/.cursor/mcp.json
Windows: %USERPROFILE%\.cursor\mcp.json
Linux: ~/.cursor/mcp.json
{
"mcpServers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Windsurf (Codeium)
Windsurf supports MCP through its configuration.
Configuration File Location
macOS: ~/.codeium/windsurf/mcp_config.json
Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json
Linux: ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
Zed Editor
Zed has native MCP support. Configure in Zed settings.
Configuration
Edit ~/.config/zed/settings.json:
{
"language_models": {
"mcp_servers": {
"pdfgenstudio": {
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
}
}
}
Continue (VS Code/JetBrains Extension)
Continue supports MCP servers for enhanced AI capabilities.
Configuration
Edit ~/.continue/config.json:
{
"mcpServers": [
{
"name": "pdfgenstudio",
"command": "npx",
"args": ["-y", "@pdfgenstudio/mcp"],
"env": {
"PDFGENSTUDIO_API_KEY": "your-api-key-here"
}
}
]
}
HTTP/SSE Mode (For Custom Clients)
For clients that support HTTP-based MCP connections:
Start the Server
# Start HTTP server
npx @pdfgenstudio/mcp --transport http --port 3100
Endpoints
- HTTP Streaming:
http://localhost:3100/mcp - SSE:
http://localhost:3100/sse - Health Check:
http://localhost:3100/health
Example Client Connection
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StreamableHTTPClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp.js";
const transport = new StreamableHTTPClientTransport(
new URL("http://localhost:3100/mcp")
);
const client = new Client({
name: "my-client",
version: "1.0.0",
});
await client.connect(transport);
Troubleshooting
Common Issues
-
"API key not configured"
- Ensure
PDFGENSTUDIO_API_KEYis set in theenvsection - Check for typos in the environment variable name
- Ensure
-
"Command not found" (npx)
- Make sure Node.js 18+ is installed
- Try using the full path to npx:
/usr/local/bin/npx
-
Server not connecting
- Restart your IDE/application after configuration changes
- Check the logs for error messages
- Verify the path is correct for local installations
-
Windows path issues
- Use forward slashes
/or escaped backslashes\\in paths - Use absolute paths
- Use forward slashes
Debug Mode
Run with debug logging:
DEBUG=* npx @pdfgenstudio/mcp
Verify Installation
# Test the server directly
npx @pdfgenstudio/mcp --help
# Test with MCP Inspector
npx fastmcp inspect /path/to/pdfgenstudio-mcp/src/index.ts
Available Tools
Template Tools
| Tool | Description |
|---|---|
render_template |
Render a template to PDF/PNG/JPG with data injection |
render_template_image |
Render template and return viewable image |
JSON Tools
| Tool | Description |
|---|---|
render_json |
Convert JSON design document to PDF/image |
render_json_image |
Render JSON and return viewable image |
HTML Tools
| Tool | Description |
|---|---|
render_html |
Convert HTML/CSS to PDF/image |
render_html_image |
Render HTML and return viewable image |
URL Tools
| Tool | Description |
|---|---|
render_url |
Screenshot webpage as PDF/image |
render_url_image |
Screenshot and return viewable image |
Management Tools
| Tool | Description |
|---|---|
list_templates |
List all saved templates |
get_template |
Get template details |
get_template_schema |
Get modifiable template elements |
Resources
The MCP server provides these resources:
| URI | Description |
|---|---|
pdfgenstudio://templates |
List of all templates |
pdfgenstudio://templates/{id} |
Specific template details |
pdfgenstudio://docs/api |
API documentation |
pdfgenstudio://config |
Current configuration status |
Prompts
Pre-built prompts for common tasks:
| Prompt | Description |
|---|---|
generate-invoice |
Generate invoice PDFs |
generate-report |
Create report documents |
capture-webpage |
Screenshot webpages |
use-template |
Render templates with data |
html-to-pdf |
Convert HTML to PDF |
Example Usage
Render a Template
Use the render_template tool to generate a PDF from template ID "abc123"
with the following data: {"name": "John Doe", "amount": "$500"}
Screenshot a Webpage
Capture a full-page screenshot of https://example.com as PNG
Generate HTML Report
Convert this HTML to PDF:
<html>
<body>
<h1>Monthly Report</h1>
<p>Content here...</p>
</body>
</html>
API Reference
For detailed API documentation, visit docs.pdfgenstudio.com
Development
# Install dependencies
npm install
# Build
npm run build
# Watch mode
npm run dev
# Test with MCP CLI
npm run test
# Inspect with MCP Inspector
npm run inspect
License
MIT License - see LICENSE for details.
Support
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。