Skip to main content
Glama

Instantly MCP Server

QUICK-START.md6.08 kB
# ⚡ **QUICK START GUIDE** ## Deploy & Test Instantly MCP Server in 10 Minutes Get your streaming HTTP transport running and tested before instantly.ai integration. --- ## 🚀 **OPTION 1: RAILWAY DEPLOYMENT (RECOMMENDED)** ### **Step 1: One-Command Deploy** ```bash # Set your API key export INSTANTLY_API_KEY="your_api_key_here" # Deploy to Railway ./deploy-staging.sh ``` **That's it!** The script will: - ✅ Install Railway CLI if needed - ✅ Create and deploy project - ✅ Set environment variables - ✅ Run comprehensive tests - ✅ Provide staging URL ### **Step 2: Get Your URL** After deployment, you'll get a URL like: ``` https://instantly-mcp-staging-production.up.railway.app ``` ### **Step 3: Test Immediately** ```bash # Quick health check curl https://your-staging-url.railway.app/health # Test MCP endpoint curl -X POST https://your-staging-url.railway.app/mcp \ -H "Content-Type: application/json" \ -H "x-api-key: YOUR_API_KEY" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' ``` --- ## 🚀 **OPTION 2: MANUAL RAILWAY SETUP** ### **Step 1: Install Railway CLI** ```bash npm install -g @railway/cli railway login ``` ### **Step 2: Deploy** ```bash # Build project npm run build # Create Railway project railway new instantly-mcp-staging # Set environment variables railway variables set INSTANTLY_API_KEY="your_api_key" railway variables set NODE_ENV=production railway variables set TRANSPORT_MODE=http # Deploy railway up ``` ### **Step 3: Test** ```bash # Get your URL railway domain # Run tests STAGING_URL="https://your-url.railway.app" node test-staging-environment.js ``` --- ## 🚀 **OPTION 3: RENDER DEPLOYMENT** ### **Step 1: Connect Repository** 1. Go to [render.com](https://render.com) 2. Connect your GitHub repository 3. Choose "Web Service" ### **Step 2: Configure** - **Build Command**: `npm run build` - **Start Command**: `npm run start:production` - **Environment Variables**: ``` NODE_ENV=production TRANSPORT_MODE=http INSTANTLY_API_KEY=your_api_key ``` ### **Step 3: Deploy & Test** Render will provide a URL like `https://instantly-mcp.onrender.com` --- ## 🧪 **INSTANT TESTING** ### **1. Health Check (30 seconds)** ```bash curl https://your-staging-url/health # Expected: {"status":"healthy","service":"instantly-mcp"} ``` ### **2. Tools List (1 minute)** ```bash curl -X POST https://your-staging-url/mcp \ -H "Content-Type: application/json" \ -H "x-api-key: YOUR_API_KEY" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' # Expected: 22 tools listed ``` ### **3. Tool Execution (2 minutes)** ```bash curl -X POST https://your-staging-url/mcp \ -H "Content-Type: application/json" \ -H "x-api-key: YOUR_API_KEY" \ -d '{ "jsonrpc": "2.0", "id": 2, "method": "tools/call", "params": { "name": "list_accounts", "arguments": {"get_all": true} } }' # Expected: Account data returned ``` ### **4. Comprehensive Test Suite (5 minutes)** ```bash STAGING_URL="https://your-staging-url" \ INSTANTLY_API_KEY="your_api_key" \ node test-staging-environment.js ``` --- ## 📋 **INSTANT VALIDATION CHECKLIST** ### **✅ Basic Functionality** - [ ] Health endpoint returns 200 - [ ] Server info shows 22 tools - [ ] Tools list returns all tools - [ ] Tool execution works - [ ] Authentication required ### **✅ Production Readiness** - [ ] HTTPS enabled - [ ] CORS configured - [ ] Error handling works - [ ] Performance acceptable - [ ] Monitoring ready ### **✅ Client Compatibility** - [ ] n8n HTTP requests work - [ ] Browser fetch API works - [ ] Postman/curl work - [ ] Local stdio still works --- ## 🎯 **SUCCESS INDICATORS** ### **You'll Know It's Working When:** 1. **Health check returns**: `{"status":"healthy"}` 2. **Tools list shows**: 22 tools available 3. **Tool execution**: Returns actual data 4. **Authentication**: Rejects requests without API key 5. **Test suite**: All tests pass ### **Ready for Instantly.ai When:** - ✅ All tests passing - ✅ Performance under 2 seconds - ✅ No errors in logs - ✅ CORS working for instantly.ai - ✅ Authentication enforced --- ## 📞 **INSTANT SUPPORT** ### **If Something Goes Wrong:** #### **Deployment Issues** ```bash # Check Railway logs railway logs # Check build status railway status # Redeploy railway up --detach ``` #### **API Issues** ```bash # Test health first curl https://your-url/health # Check API key echo $INSTANTLY_API_KEY # Test without auth (should fail) curl -X POST https://your-url/mcp \ -H "Content-Type: application/json" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' ``` #### **Performance Issues** ```bash # Check response time time curl https://your-url/health # Monitor logs railway logs --tail ``` --- ## 🚀 **NEXT STEPS AFTER SUCCESS** ### **1. Share with Instantly.ai Team** ```markdown ✅ Staging URL: https://your-staging-url.railway.app ✅ All tests passing ✅ Ready for production deployment ✅ See INSTANTLY-TEAM-HANDOFF.md for details ``` ### **2. Prepare for Production** - Domain configuration: `instantly.ai/mcp` - SSL certificate setup - Production API key - Monitoring configuration ### **3. Go Live** Once Instantly.ai team configures DNS: ```bash # Test production endpoint curl https://instantly.ai/health curl -X POST https://instantly.ai/mcp \ -H "Content-Type: application/json" \ -H "x-api-key: YOUR_API_KEY" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' ``` --- ## 🎉 **YOU'RE DONE!** **In just 10 minutes, you have:** - ✅ Deployed streaming HTTP MCP server - ✅ Tested all functionality - ✅ Validated client compatibility - ✅ Prepared for instantly.ai integration **Your staging environment is ready to share with the Instantly.ai team for production deployment!** 🚀 --- *Need help? Check the detailed guides:* - 📖 **STAGING-DEPLOYMENT.md** - Comprehensive deployment guide - 🧪 **MCP-CLIENT-TESTING.md** - Client testing procedures - 📋 **INSTANTLY-TEAM-HANDOFF.md** - Production handoff document

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/bcharleson/Instantly-MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server