PDF Generator API MCP Server
An MCP server that enables interaction with the PDF Generator API for automated document generation and management. It supports both stdio and HTTP transports, allowing AI models to create and handle PDFs using OpenAPI v4 specifications.
README
AR API MCP Server
Model Context Protocol (MCP) server for the PDF Generator API, generated from the OpenAPI v4 specification.
Overview
This server supports both stdio and HTTP transports:
- Stdio mode (default): For local MCP client integration (Claude Desktop, Claude Code, Cline, etc.)
- HTTP mode: For production deployment, marketplaces, and multiple clients
Quick Start
Local Development (Stdio Mode)
Install and build:
npm install
npm run build
MCP Client Configuration Example:
{
"mcpServers": {
"ar-api": {
"command": "node",
"args": ["/path/to/mcp-server-v4/build/index.js"],
"env": {
"API_BASE_URL": "https://us1.pdfgeneratorapi.com/api/v4",
"BEARER_TOKEN_JWT": "your-jwt-token-here"
}
}
}
}
Note: Replace /path/to/mcp-server-v4 with your actual path and your-jwt-token-here with your PDF Generator API JWT token.
Common config locations:
- Claude Desktop: Settings → Developer → Edit Config
- Claude Code (CLI):
~/.claude/mcp_config.json - Cline/Roo-Codeium:
.vscode/mcp_config.json - Continue:
~/.continue/config.json
Production Deployment (HTTP Mode)
Run locally:
npm install
npm run start:http
# Server runs on http://localhost:3000
# MCP endpoint: http://localhost:3000/mcp
# Health check: http://localhost:3000/health
Docker:
docker compose -f deploy/docker-compose.yml up -d
# MCP endpoint: http://localhost:3001/mcp
Authentication
HTTP Mode: Pass the Bearer token in the Authorization header with each request:
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Authorization: Bearer your-jwt-token-here" \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize",...}'
Stdio Mode: Pass the Bearer token via the BEARER_TOKEN_JWT environment variable in your MCP client config.
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
API_BASE_URL |
No | https://us1.pdfgeneratorapi.com/api/v4 |
PDF Generator API base URL |
PORT |
No | 3000 |
Server port (HTTP mode only) |
BEARER_TOKEN_JWT |
No | - | JWT token (stdio mode only) |
LOG_LEVEL |
No | info |
Logging level |
Create a .env file in the repo root:
cp .env.example .env
JWT Token Generation
Generate a JWT token using your PDF Generator API credentials:
const jwt = require('jsonwebtoken');
const token = jwt.sign(
{ iss: 'YOUR_WORKSPACE_ID', sub: 'YOUR_WORKSPACE_IDENTIFIER' },
'YOUR_SECRET_KEY',
{ algorithm: 'HS256', expiresIn: '1h' }
);
See: https://docs.pdfgeneratorapi.com/v4#section/Authentication/Creating-a-JWT
Regenerating
To regenerate after OpenAPI spec changes:
# Place the updated spec at docs/apiv4.json, then:
./scripts/generate-mcp.sh
# Or specify a custom input:
./scripts/generate-mcp.sh --input /path/to/apiv4.json
Note: After regeneration, custom patches may need to be re-applied.
Testing
Test stdio mode:
npm run build
BEARER_TOKEN_JWT="your-token" echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | npm run start:stdio
Test HTTP mode:
npm run start:http
curl http://localhost:3000/health
See POSTMAN_TESTING.md for detailed testing instructions with Postman.
Resources
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。