WyreUP MCP Server

A production-ready Model Context Protocol (MCP) server that transforms webhook endpoints from automation platforms (n8n, Make.com, Zapier, FlowiseAI, etc.) into reliable, agent-callable tools for any MCP-compatible AI system.

Compatible with:

• 🤖 Claude Desktop (Anthropic's official desktop app)
• 🔧 Any MCP Client (VS Code extensions, custom applications, etc.)
• 🌐 Future MCP Services (growing ecosystem of MCP-enabled tools)
• 📱 Custom Integrations (build your own MCP client)

📦 NPM Package: wyreup-mcp 🔗 GitHub Repository: tamler/wyreup-mcp

🎯 What It Does

Turn any HTTP webhook into an MCP tool that AI agents can use reliably. Define your automation endpoints in a simple JSON manifest, and the server handles authentication, retries, rate limiting, health monitoring, and error handling automatically.

✨ Key Features

🚀 Enterprise-Ready Reliability

• Smart Retry Logic: Exponential backoff for transient failures
• Rate Limiting: Per-tool request throttling with sliding windows
• Health Monitoring: Real-time endpoint health tracking and statistics
• Timeout Management: Configurable timeouts per tool (great for slow automations)

🔐 Flexible Authentication

• Multiple Auth Types: Header-based and JWT Bearer token authentication
• Environment Variables: Secure credential management via valueFromEnv and tokenFromEnv
• External Secrets: Store credentials in ~/.wyreup-secrets/ files
• Per-Tool Auth: Each webhook can use different authentication methods

🛠 Developer-Friendly

• JSON Schema Validation: Full input/output schema support with Zod validation
• Rich Error Handling: Structured error responses with debugging context
• Built-in Monitoring Tools: Health checks and rate limit status via MCP
• Hot Reload: Changes to manifest files are picked up automatically

🌊 Advanced HTTP Support

• All HTTP Methods: GET, POST, PUT, PATCH, DELETE support
• Binary Data: Handle file downloads and binary responses
• Streaming Ready: Architecture prepared for real-time data streams
• Content Type Detection: Automatic handling of JSON, text, and binary responses

🧠 MCP Protocol Compliance

• ✅ Uses official MCP SDK (v1.12.0+) with full STDIO and SSE support
• ✅ Per-session SSE transports with isolated tool execution
• Supports transports: STDIO and SSE
• Tool Registration: Fully schema-based using tools/list and tools/call
• JSON-RPC 2.0: Strictly compliant request/response structure

🚀 Quick Start

Installation & Setup

# Install globally from npm
npm install -g wyreup-mcp

# Or run directly with npx (recommended)
npx wyreup-mcp --init
npx wyreup-mcp --validate
npx wyreup-mcp

⚡ Ultra-Fast Setup: Simplified Format

NEW: Get started in under 2 minutes with the simplified webhook format! Perfect for rapid prototyping and low-code users.

{
  "tools": [
    {
      "name": "get_quote",
      "webhook": "https://wyreup.com/tools-mock/random-quote"
    }
  ]
}

That's it! The system automatically infers:

• ✅ Description: "Forward to random quote webhook"
• ✅ Method: POST (default for webhooks)
• ✅ Input Schema: Empty object {}
• ✅ Output Schema: Auto-detects JSON, text, binary, or streaming responses

Supported Response Types:

• 📄 JSON - Auto-parsed for structured data
• 📝 Text/HTML - Plain text responses
• 🖼️ Binary - Images, PDFs, files (base64 encoded)
• 🌊 Streaming - Real-time event streams

Test immediately:

echo '{
  "tools": [
    {
      "name": "get_quote",
      "webhook": "https://wyreup.com/tools-mock/random-quote"
    }
  ]
}' > wyreup-simple.json

echo '{ "jsonrpc": "2.0", "id": "call-1", "method": "tools/call", "params": { "name": "get_quote", "arguments": {} } }' | npx wyreup-mcp --config wyreup-simple.json --transport stdio --debug

Simplified Format Syntax:

{
  "tools": [
    {
      "name": "your_tool_name",
      "webhook": "https://your-webhook-url.com/endpoint"
    }
  ]
}

Mixed Format Support: You can combine simplified and full format tools in the same configuration file for maximum flexibility.

🚀 Instant Testing with Live Endpoints

Skip the setup and test immediately with our live demo endpoints at wyreup.com:

# Create a quick test configuration
echo '{
  "tools": [
    {
      "name": "get_quote",
      "description": "Get an inspirational quote",
      "url": "https://wyreup.com/tools-mock/random-quote",
      "method": "GET",
      "input": {},
      "output": {
        "type": "object",
        "properties": {
          "quote": { "type": "string" }
        }
      }
    },
    {
      "name": "get_time",
      "description": "Get current UTC time",
      "url": "https://wyreup.com/tools-mock/current-time",
      "method": "GET",
      "auth": {
        "type": "header",
        "name": "X-API-Key",
        "value": "mock-secret-key"
      },
      "output": {
        "type": "object",
        "properties": {
          "time": { "type": "string" }
        }
      }
    }
  ]
}' > test-wyreup.json

# Test immediately
npx wyreup-mcp --validate --config test-wyreup.json
npx wyreup-mcp --config test-wyreup.json

Or use the comprehensive example configuration:

# Download the full example config with all advanced features
curl -o wyreup-example.json https://raw.githubusercontent.com/tamler/wyreup-mcp/main/wyreup-example.json

# Test all features
npx wyreup-mcp --validate --config wyreup-example.json
npx wyreup-mcp --config wyreup-example.json

# Run the full SSE test suite
npm run test:sse
npm run test:sse-client

Available Test Endpoints:

• 🗨️ Random Quote - GET /tools-mock/random-quote (no auth)
• ⏰ Current Time - GET /tools-mock/current-time (API key: mock-secret-key)
• 🔄 Echo Service - POST /tools-mock/echo (API key: test-api-key-12345)
• ❌ Error Testing - POST /tools-mock/error (always returns 500)
• 🖼️ Image Generation - GET /tools-mock/generate-image (returns base64 PNG)
• 🎵 Audio Generation - GET /tools-mock/generate-audio (returns base64 WAV with bytebeat music!)

Default Configuration

The default wyreup.json demonstrates both simplified and full formats:

{
  "tools": [
    {
      "name": "get_quote",
      "webhook": "https://wyreup.com/tools-mock/random-quote"
    },
    {
      "name": "echo_message",
      "description": "Echoes back the JSON body. Expects 'message' and 'detail'.",
      "url": "https://wyreup.com/tools-mock/echo",
      "public": false,
      "input": { "message": "string", "detail": "string" },
      "output": {
        "received_message": "string",
        "received_detail": "string",
        "timestamp": "string"
      },
      "method": "POST",
      "auth": {
        "type": "header",
        "name": "X-API-Key",
        "value": "test-api-key-12345"
      }
    }
  ]
}

For comprehensive examples with all advanced features, see wyreup-example.json.

Advanced Configuration Features

For production deployments, you can add advanced features:

{
  "tools": [
    {
      "name": "advanced_tool",
      "description": "Production-ready tool with all features",
      "url": "https://your-automation-platform.com/webhook/endpoint",
      "method": "POST",
      "timeout": 30000,
      "maxRetries": 3,
      "retryDelay": 1000,
      "rateLimit": {
        "requests": 10,
        "window": 60000
      },
      "input": {
        "type": "object",
        "properties": {
          "url": { "type": "string", "description": "URL to summarize" },
          "max_words": { "type": "integer", "default": 150 }
        },
        "required": ["url"]
      },
      "output": {
        "type": "object",
        "properties": {
          "summary": { "type": "string" },
          "word_count": { "type": "integer" }
        }
      },
      "auth": {
        "type": "header",
        "name": "X-API-Key",
        "valueFromEnv": "SUMMARIZE_API_KEY"
      }
    }
  ]
}

Connect to MCP Clients

Claude Desktop: Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "wyreup": {
      "command": "npx",
      "args": ["wyreup-mcp"],
      "cwd": "/path/to/your/project"
    }
  }
}

Other MCP Clients:

• VS Code Extensions: Use MCP extension settings
• Custom Applications: Connect via stdio transport on any platform
• Server Deployments: Use SSE transport mode (--transport sse --port 3333)

The server implements the standard MCP protocol, so it works with any compliant client.

🔧 Configuration Reference

Tool Properties

Property	Type	Description
name	string	Required. Unique tool identifier
description	string	Required. Human-readable description
url	string	Required. Full webhook URL
method	string	HTTP method (default: POST)
timeout	number	Request timeout in milliseconds (default: 30000)
maxRetries	number	Maximum retry attempts (default: 3)
retryDelay	number	Base retry delay in ms (default: 1000)
rateLimit	object	Rate limiting configuration
input	object	JSON Schema for input validation
output	object	JSON Schema for output description
auth	object	Authentication configuration
authFrom	object	External authentication source

Authentication Types

Header Authentication:

{
  "auth": {
    "type": "header",
    "name": "Authorization",
    "value": "Bearer secret-token"
  }
}

Environment Variable:

{
  "auth": {
    "type": "header",
    "name": "X-API-Key",
    "valueFromEnv": "API_KEY"
  }
}

JWT Bearer:

{
  "auth": {
    "type": "jwt",
    "tokenFromEnv": "JWT_TOKEN"
  }
}

External Secrets:

{
  "authFrom": {
    "user": "production-user"
  }
}

Create ~/.wyreup-secrets/production-user.json:

{
  "tool_name": {
    "type": "header",
    "name": "Authorization",
    "value": "Bearer your-secret-token"
  }
}

Rate Limiting

{
  "rateLimit": {
    "requests": 30, // Max requests
    "window": 60000 // Time window in ms
  }
}

🖥 Usage

Command Line Options

# Validation
npx wyreup-mcp --validate
npx wyreup-mcp --validate --config wyreup-example.json

# Server modes
npx wyreup-mcp                          # STDIO mode (default, used by Claude Desktop and local tools)
npx wyreup-mcp --transport sse         # SSE mode for web clients (experimental)

# Development
npx wyreup-mcp --debug                  # Enable debug logging
npx wyreup-mcp --init                   # Create sample manifest

# SSE Testing (experimental)
npm run test:sse         # Starts local SSE server with demo tools
npm run test:sse-client  # Sends a tool call over SSE transport

Built-in Monitoring

These tools are automatically available and do not require configuration:

• health-check: Test individual webhook endpoints
• health-status: Get success rates and performance metrics
• rate-limit-status: Monitor rate limiting usage

Example: Check health of all tools

echo '{"jsonrpc": "2.0", "method": "tools/call", "params": {"name": "health-status", "arguments": {}}, "id": 1}' | npx wyreup-mcp

🧪 SSE Transport Notes

To test SSE locally:

1. Start the server with --transport sse
2. Your client must subscribe to /sse (EventSource)
3. Wait for the endpoint event to get the sessionId
4. Send POST requests to /messages?sessionId=...
5. Each session gets its own transport and tool server instance
6. Transport is cleaned up after tool execution or connection close

🌟 Real-World Examples

Content Processing Pipeline

{
  "name": "process_article",
  "description": "Extract, summarize, and analyze web articles",
  "url": "https://n8n.company.com/webhook/process-article",
  "timeout": 45000,
  "maxRetries": 2,
  "input": {
    "type": "object",
    "properties": {
      "url": { "type": "string", "description": "Article URL" },
      "analysis_type": {
        "type": "string",
        "enum": ["summary", "sentiment", "keywords", "full"],
        "default": "summary"
      }
    },
    "required": ["url"]
  }
}

Slack Notifications

{
  "name": "notify_team",
  "description": "Send notifications to team channels",
  "url": "https://hooks.zapier.com/hooks/catch/slack-notify",
  "timeout": 15000,
  "rateLimit": { "requests": 30, "window": 60000 },
  "input": {
    "type": "object",
    "properties": {
      "channel": { "type": "string", "description": "#channel-name" },
      "message": { "type": "string" },
      "priority": {
        "type": "string",
        "enum": ["low", "normal", "high"],
        "default": "normal"
      }
    },
    "required": ["channel", "message"]
  }
}

Document Generation

{
  "name": "generate_report",
  "description": "Generate PDF reports from data",
  "url": "https://make.com/webhook/generate-pdf",
  "timeout": 60000,
  "input": {
    "type": "object",
    "properties": {
      "template": { "type": "string" },
      "data": { "type": "object" },
      "format": { "type": "string", "enum": ["pdf", "docx"] }
    }
  },
  "output": {
    "binary": true,
    "contentType": "application/pdf"
  }
}

🚀 Future Roadmap

Near-Term Enhancements

• Webhook Signature Verification: Validate incoming webhook signatures for security
• Response Caching: Cache responses for identical requests to reduce API calls
• Connection Pooling: Optimize HTTP connections for high-throughput scenarios
• Prometheus Metrics: Export detailed metrics for production monitoring

Advanced Features

• Auto-Discovery: Automatically detect and configure tools from popular platforms
• Visual Tool Builder: Web UI for creating and testing webhook configurations
• Team Management: Multi-user authentication and permission systems
• Cloud Registry: Share and discover webhook tools across teams

Platform Integrations

• Native Platform Support: Pre-built templates for n8n, Make.com, Zapier, etc.
• OAuth Integration: Simplified authentication flows for supported platforms
• Batch Operations: Support for bulk webhook calls and parallel processing
• Real-time Subscriptions: WebSocket support for live data feeds

🤝 Contributing

Contributions welcome! Please visit our GitHub repository to:

• Report Issues: GitHub Issues
• Submit Pull Requests: GitHub PRs
• View Source: GitHub Repository

Key areas for contribution:

• Platform Templates: Pre-configured tool definitions for popular automation platforms
• Performance Optimization: Enhance retry logic, connection handling, and response processing
• Security Features: Webhook signature verification, request sanitization
• Monitoring & Observability: Enhanced metrics, logging, and debugging tools

📄 License

MIT License - Build amazing automations for AI agents!

---

Transform your automation workflows into AI-ready tools in minutes, not hours.