@agentlify/mcp-server 2.0.0

package/docs/quickstart.md

@@ -0,0 +1,242 @@
+# Agentlify 5-Minute Quickstart
+
+Get started with intelligent AI routing in under 5 minutes.
+
+---
+
+## What You'll Do
+
+1. Create account & router (2 min)
+2. Update your code (1 min)
+3. Make your first request (1 min)
+4. View cost savings (1 min)
+
+**Total: 5 minutes to 40-60% cost savings!**
+
+---
+
+## Step 1: Create Account & Router (2 min)
+
+### 1.1 Sign Up
+
+Visit [agentlify.app](https://agentlify.app) and create your free account.
+
+### 1.2 Create Your First Router
+
+Click "Create Router" and choose your optimization priority:
+
+- **Cost-Optimized** 💰 - Minimize costs (recommended for most)
+- **Quality-Optimized** 🎯 - Best performance
+- **Speed-Optimized** ⚡ - Lowest latency
+
+Router is created instantly with access to 100+ models.
+
+### 1.3 Get Credentials
+
+1. **API Key**: Go to Settings → API Keys → Generate (starts with `mp_`)
+2. **Router ID**: Copy from your router's page
+
+---
+
+## Step 2: Update Your Code (1 min)
+
+### Python
+
+```python
+from openai import OpenAI  # Keep using OpenAI SDK!
+
+# Just change these two lines:
+client = OpenAI(
+    api_key="mp_your_api_key",
+    base_url="https://agentlify.co/api/router/your_router_id"
+)
+
+# Everything else works the same!
+response = client.chat.completions.create(
+    messages=[{"role": "user", "content": "Hello!"}]
+)
+
+print(response.choices[0].message.content)
+```
+
+### JavaScript/TypeScript
+
+```javascript
+const { OpenAI } = require('openai'); // Keep using OpenAI SDK!
+
+// Just change these two lines:
+const client = new OpenAI({
+  apiKey: 'mp_your_api_key',
+  baseURL: 'https://agentlify.co/api/router/your_router_id',
+});
+
+// Everything else works the same!
+const response = await client.chat.completions.create({
+  messages: [{ role: 'user', content: 'Hello!' }],
+});
+
+console.log(response.choices[0].message.content);
+```
+
+---
+
+## Step 3: Make Your First Request (1 min)
+
+Run your code! Agentlify will:
+
+1. ✅ Automatically select the best model
+2. ✅ Route your request intelligently
+3. ✅ Return the response
+4. ✅ Track costs and performance
+
+### Example Response:
+
+```javascript
+{
+  choices: [{
+    message: { content: "Hello! How can I help you today?" }
+  }],
+  _meta: {
+    cost: 0.0008,           // Actual cost in USD
+    modelUsed: "gpt-4o-mini", // Which model was selected
+    latency: 450,           // Response time in ms
+    requestId: "req_abc123"
+  }
+}
+```
+
+---
+
+## Step 4: View Cost Savings (1 min)
+
+### Dashboard Analytics
+
+Go to your router's dashboard to see:
+
+- 💰 **Total cost** vs. OpenAI-only cost
+- 📊 **Savings percentage** (typically 40-60%)
+- 📈 **Request volume** over time
+- 🤖 **Model distribution** (which models are being used)
+
+### Use MCP Tools (Optional)
+
+If you're using an AI assistant with MCP:
+
+```
+User: "Show my usage for the last 7 days"
+
+AI Assistant:
+Total cost: $12.45
+Requests: 1,234
+Average cost per request: $0.0101
+Top model: gpt-4o-mini (65%)
+Savings vs OpenAI: 52% ($13.58 saved)
+```
+
+---
+
+## What's Next?
+
+You're all set! Here's what you can explore:
+
+### 📚 Learn More
+
+- [OpenAI Migration Guide](./migration-openai.md) - Detailed migration
+- [Router Configuration](./router-configuration.md) - Advanced settings
+- [Cost Optimization](./cost-optimization.md) - Save even more
+
+### 🛠️ Optimize Further
+
+Use MCP tools in your AI assistant:
+
+```
+"Optimize my router for cost savings"
+"Compare GPT-4 vs Claude 3.5 Sonnet"
+"Generate production code for my router"
+```
+
+### 🎯 Common Use Cases
+
+**Chatbot:**
+
+```javascript
+const response = await client.chat.completions.create({
+  messages: conversationHistory,
+  stream: true, // Streaming works!
+});
+```
+
+**Function Calling:**
+
+```javascript
+const response = await client.chat.completions.create({
+  messages: [{ role: 'user', content: 'Get weather in NYC' }],
+  tools: [{ type: 'function', function: weatherTool }],
+});
+```
+
+**JSON Mode:**
+
+```javascript
+const response = await client.chat.completions.create({
+  messages: [{ role: 'user', content: 'Generate JSON data' }],
+  response_format: { type: 'json_object' },
+});
+```
+
+---
+
+## Key Features
+
+✅ **Zero code changes** - Use existing OpenAI SDK  
+✅ **Automatic optimization** - Router selects best model  
+✅ **40-60% cost savings** - Intelligent routing  
+✅ **100+ models** - OpenAI, Anthropic, Google, more  
+✅ **Real-time analytics** - Track every request  
+✅ **Automatic fallbacks** - Improved reliability
+
+---
+
+## Environment Variables (Best Practice)
+
+Create `.env` file:
+
+```bash
+AGENTLIFY_API_KEY=mp_your_api_key_here
+AGENTLIFY_ROUTER_ID=your_router_id_here
+```
+
+Update your code:
+
+```javascript
+const client = new OpenAI({
+  apiKey: process.env.AGENTLIFY_API_KEY,
+  baseURL: `https://agentlify.co/api/router/${process.env.AGENTLIFY_ROUTER_ID}`,
+});
+```
+
+---
+
+## Troubleshooting
+
+### "Invalid API Key"
+
+- Make sure you're using your Agentlify key (`mp_xxx`), not OpenAI key
+- Check Settings → API Keys in dashboard
+
+### "Router not found"
+
+- Verify Router ID in baseURL matches your router
+- Check router is active in dashboard
+
+### Need Help?
+
+- **Docs**: [agentlify.app/docs](https://agentlify.app/docs)
+- **Support**: support@agentlify.app
+- **AI Assistant**: Use MCP tools for instant help
+
+---
+
+**That's it!** You're now using Agentlify for intelligent AI routing. 🎉
+
+**Next recommended reading:** [OpenAI Migration Guide](./migration-openai.md)