architectgbt-mcp 0.2.1 → 0.2.3

package/README.md

@@ -1,189 +1,162 @@
 # ArchitectGBT MCP Server
 
-AI model recommendation engine for Cursor, Claude Desktop, and Windsurf.
+**Model Context Protocol (MCP) server for AI model recommendations and code templates.**
 
-Get instant AI model recommendations without leaving your IDE.
+Get instant AI model recommendations for your projects directly in Cursor IDE or Claude Desktop. Compare **50+ models** from OpenAI, Anthropic, Google, Meta, and Mistral with pricing, capabilities, and production-ready code templates.
 
-## Features
-
-- 🎯 **Smart Recommendations** - Get the best AI model for your use case
-- 📝 **Code Templates** - Production-ready integration code
-- 📊 **Model Database** - Compare 50+ AI models with pricing
-
-## Installation
+[![npm version](https://img.shields.io/npm/v/architectgbt-mcp.svg)](https://www.npmjs.com/package/architectgbt-mcp)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-### For Cursor
+## Features
 
-Add to your `~/.cursor/mcp.json` (or `C:\Users\YourName\.cursor\mcp.json` on Windows):
+### 🎯 AI Model Recommendations
+- **Personalized suggestions** based on your project description
+- **Budget optimization** (low/medium/high/unlimited)
+- **Priority matching** (cost/speed/quality/balanced)
+- **Smart analysis** with reasoning, pros/cons, and alternatives
+- **Cost estimates** with realistic token usage calculations
 
-```json
-{
-  "mcpServers": {
-    "architectgbt": {
-      "command": "npx",
-      "args": ["-y", "architectgbt-mcp"]
-    }
-  }
-}
-```
+### 📊 Model Database
+- **50+ AI models** across all major providers
+- **Real-time pricing** per 1M tokens (input/output)
+- **Detailed specs** (context windows, capabilities, speed rankings)
+- **Provider filtering** (OpenAI, Anthropic, Google, Meta, Mistral)
 
-**Important:** Restart Cursor completely after adding the configuration.
+### 💻 Code Templates
+- **Production-ready integrations** for Anthropic, OpenAI, and Google
+- **TypeScript and Python** support
+- **Complete examples** with installation, env setup, and usage
+- **Best practices** including error handling and streaming
 
-### For Claude Desktop
+## Quick Start
 
-Add to your Claude Desktop config (`%APPDATA%\Claude\claude_desktop_config.json` on Windows):
+### For Cursor IDE
 
+1. Create `.cursor/mcp.json` in your project:
 ```json
 {
   "mcpServers": {
     "architectgbt": {
       "command": "npx",
-      "args": ["architectgbt-mcp"]
+      "args": ["-y", "architectgbt-mcp@latest"]
     }
   }
 }
 ```
 
-### For Windsurf
+2. Restart Cursor
 
-Add to your MCP configuration:
+3. Ask Claude:
+   - "Recommend an AI model for a customer support chatbot"
+   - "Show me all available AI models"
+   - "Give me Claude Sonnet code in TypeScript"
 
-```json
-{
-  "mcpServers": {
-    "architectgbt": {
-      "command": "npx",
-      "args": ["architectgbt-mcp"]
-    }
-  }
-}
-```
+## Pricing
 
-## Available Tools
+| Tier | Recommendations | Cost | API Key |
+|------|----------------|------|---------|
+| **Free** | 3/day | $0 | Not needed |
+| **Pro** | Unlimited | $15/month | Required |
 
-### `list_models`
+[Get your API key →](https://architectgbt.com/dashboard/settings)
 
-List available AI models with pricing information.
+## Tools
 
-**Example prompts:**
-- "List all Anthropic models"
-- "Show me all models"
-- "What OpenAI models are available?"
+### `list_models`
+Lists all 50+ available AI models with pricing.
 
-**Example output:**
-```
-## 📊 Available AI Models
-
-| Model | Provider | Input $/1M | Output $/1M |
-|-------|----------|------------|-------------|
-| Claude Haiku 4.5 | Anthropic | $0.80 | $4.00 |
-| Claude Opus 4.5 | Anthropic | $15.00 | $75.00 |
-| Claude Sonnet 4.5 | Anthropic | $3.00 | $15.00 |
-| GPT-4o | OpenAI | $2.50 | $10.00 |
-| GPT-4o mini | OpenAI | $0.15 | $0.60 |
-...
-
-*Showing 10 models. Use `get_ai_recommendation` for personalized suggestions.*
 ```
+User: "Show me all AI models"
 
-### `get_code_template`
-
-Get production-ready integration code for any AI model.
+Response:
+Available AI Models (50 total)
+======================================================================
 
-**Supported providers:** Anthropic (Claude), OpenAI (GPT), Google (Gemini)  
-**Languages:** TypeScript, Python
-
-**Example prompts:**
-- "Give me a TypeScript template for Claude"
-- "Python code for OpenAI GPT-4"
-- "Gemini integration in TypeScript"
-
-**Example output:**
-```typescript
-// ✅ Installation
-npm install @anthropic-ai/sdk
-
-// 🔑 Environment Variables
-ANTHROPIC_API_KEY=your_api_key_here
-
-// 📦 Code
-import Anthropic from '@anthropic-ai/sdk';
-
-const client = new Anthropic({
-  apiKey: process.env.ANTHROPIC_API_KEY,
-});
+Anthropic:
+  • Claude Haiku 4.5      $  0.80 / $  4.00 (in/out per 1M tokens)
+  • Claude Sonnet 4.5     $  3.00 / $ 15.00 (in/out per 1M tokens)
+  ...
+```
 
-const message = await client.messages.create({
-  model: 'claude-3-5-sonnet-20241022',
-  max_tokens: 1024,
-  messages: [
-    { role: 'user', content: 'Hello, Claude!' }
-  ],
-});
+**Parameters:**
+- `provider` (optional): Filter by "OpenAI" | "Anthropic" | "Google" | "Meta" | "Mistral"
+- `limit` (optional): Number of models (default: 50)
 
-console.log(message.content);
+### `get_ai_recommendation`
+Get personalized AI model recommendations.
 
-// 💡 Usage Tips
-// - Use streaming for real-time responses
-// - Add system prompts for better control
-// - Handle rate limits with retries
 ```
+User: "I need an AI for a chatbot handling 10k requests/day"
 
-### `get_ai_recommendation`
+Response:
+🎯 AI Model Recommendation — Analysis Complete!
 
-Get personalized AI model recommendations based on your project description.
+✨ TOP MATCH (94% match)
 
-**Note:** Requires authentication. For full recommendations with cost analysis and reasoning, visit [architectgbt.com](https://architectgbt.com) and sign up for a free account.
+Claude Haiku 4.5
+Provider: Anthropic
+Estimated Cost: $0.0240
+Context Window: 200,000 tokens
 
-**Example prompts:**
-- "What AI model should I use for a customer support chatbot?"
-- "Recommend a model for code generation on a budget"
-- "Best model for document analysis with 100K context?"
+💡 Why this model?
+Perfect for customer support with fast responses, strong reasoning...
 
-**Example response:**
+✅ Pros:
+  • Extremely fast (sub-second)
+  • Cost-effective at scale
+  ...
 ```
-❌ Authentication Required
 
-The ArchitectGBT API requires authentication. To get AI model recommendations:
+**Parameters:**
+- `prompt` (required): Your project description
+- `budget` (optional): "low" | "medium" | "high" | "unlimited"
+- `priority` (optional): "cost" | "speed" | "quality" | "balanced"
 
-1. Visit https://architectgbt.com
-2. Sign up for a free account (free tier available!)
-3. Use the website directly for personalized recommendations
+### `get_code_template`
+Get production-ready code templates.
 
-Alternatively, you can:
-- Use `list_models` to browse available models
-- Use `get_code_template` to get integration code for any model
+**Parameters:**
+- `provider` (required): "anthropic" | "openai" | "google"
+- `language` (required): "typescript" | "python"
 
-For your query: "customer support chatbot"
-I recommend visiting the website for a personalized analysis with cost estimates and reasoning.
-```
+## Installation Options
 
-## Development
+### Option 1: Cursor IDE (Recommended)
 
-```bash
-# Install dependencies
-npm install
+See Quick Start above.
 
-# Run in development mode
-npm run dev
+### Option 2: Claude Desktop
 
-# Build for production
-npm run build
+**macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
+**Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
 
-# Start production server
-npm start
+```json
+{
+  "mcpServers": {
+    "architectgbt": {
+      "command": "npx",
+      "args": ["-y", "architectgbt-mcp@latest"],
+      "env": {
+        "ARCHITECTGBT_API_KEY": "your_key_here"
+      }
+    }
+  }
+}
 ```
 
-## Local Testing
+### Option 3: Unlimited Access (API Key)
 
-To test locally before publishing, update your MCP config to point to the built file:
+Add your API key to `.cursor/mcp.json`:
 
 ```json
 {
   "mcpServers": {
     "architectgbt": {
-      "command": "node",
-      "args": ["C:/Users/pravi/workspacenew/architectgbt-mcp/dist/index.js"]
+      "command": "npx",
+      "args": ["-y", "architectgbt-mcp@latest"],
+      "env": {
+        "ARCHITECTGBT_API_KEY": "agbt_your_key_here"
+      }
     }
   }
 }
@@ -191,15 +164,57 @@ To test locally before publishing, update your MCP config to point to the built
 
 ## Environment Variables
 
-| Variable | Description | Default |
-|----------|-------------|---------|
-| `ARCHITECTGBT_API_URL` | API base URL | `https://architectgbt.com` |
+- `ARCHITECTGBT_API_KEY` - Your API key for unlimited access (optional)
+- `ARCHITECTGBT_API_URL` - Custom API URL (default: https://architectgbt.com)
 
-## License
+## Troubleshooting
+
+### "Daily Limit Reached"
+You've used 3 free recommendations. Options:
+1. Wait for daily reset (midnight UTC)
+2. [Get API key](https://architectgbt.com/dashboard/settings) for unlimited access
 
-MIT
+### "API Key Invalid"
+1. Check key starts with `agbt_`
+2. Regenerate from [Settings](https://architectgbt.com/dashboard/settings)
+3. Ensure it's in the `env` section of your config
+
+### Models Not Showing
+1. Check internet connection
+2. Verify: https://architectgbt.com/api/models
+3. Try: `npx architectgbt-mcp@latest` manually
+
+### MCP Not Loading
+1. Node.js >= 18.0.0 required
+2. Check config JSON syntax
+3. Restart Cursor after changes
+
+## Development
+
+```bash
+git clone https://github.com/yourusername/architectgbt-mcp.git
+cd architectgbt-mcp
+npm install
+npm run build
+npm run dev
+```
 
 ## Links
 
-- [ArchitectGBT Website](https://architectgbt.com)
-- [MCP Documentation](https://modelcontextprotocol.io)
+- **Website**: [architectgbt.com](https://architectgbt.com)
+- **Docs**: [architectgbt.com/docs/mcp-integration](https://architectgbt.com/docs/mcp-integration)
+- **API Keys**: [architectgbt.com/dashboard/settings](https://architectgbt.com/dashboard/settings)
+- **NPM**: [npmjs.com/package/architectgbt-mcp](https://www.npmjs.com/package/architectgbt-mcp)
+
+## Support
+
+- **Email**: hello@architectgbt.com
+- **Issues**: [GitHub](https://github.com/yourusername/architectgbt-mcp/issues)
+
+## License
+
+MIT © ArchitectGBT
+
+---
+
+**Built with ❤️ for developers who ship fast**

package/dist/tools/get-recommendation.js

@@ -108,28 +108,87 @@ export async function handleGetRecommendation(args) {
     }
 }
 function formatRecommendation(data) {
-    const { recommendation, reasoning, alternatives, model } = data;
-    let result = `## 🎯 AI Model Recommendation\n\n`;
-    if (model) {
-        result += `### Recommended: ${model.name}\n`;
-        result += `- **Provider:** ${model.provider}\n`;
-        result += `- **Model ID:** ${model.model_id || "N/A"}\n`;
-        if (model.input_price || model.output_price) {
-            result += `- **Pricing:** $${model.input_price}/1M input, $${model.output_price}/1M output\n`;
-        }
-        if (model.context_window) {
-            result += `- **Context Window:** ${model.context_window.toLocaleString()} tokens\n`;
+    // Handle conversational/off-topic responses
+    if (data.off_topic || data.needs_clarification) {
+        let result = data.message || '';
+        if (data.questions && Array.isArray(data.questions) && data.questions.length > 0) {
+            result += `\n\n📋 To help me recommend the perfect AI model, please tell me:\n`;
+            data.questions.forEach((q, i) => {
+                result += `${i + 1}. ${q}\n`;
+            });
         }
+        return result;
+    }
+    const recommendations = data.recommendations || [];
+    if (recommendations.length === 0) {
+        return `❌ No recommendations found. Try describing your project in more detail.`;
+    }
+    let result = `🎯 AI Model Recommendation — Analysis Complete!\n`;
+    result += `${"=".repeat(70)}\n\n`;
+    // Main recommendation
+    const top = recommendations[0];
+    result += `✨ TOP MATCH (${top.match_score || top.score || 95}% match)\n\n`;
+    result += `${top.model_name || top.name}\n`;
+    result += `Provider: ${top.provider}\n`;
+    // Pricing
+    if (top.estimated_cost) {
+        const cost = top.estimated_cost;
+        result += `Estimated Cost: $${cost.total_cost_usd?.toFixed(4) || '0.0000'}\n`;
+        result += `  └─ ${cost.input_tokens?.toLocaleString() || '0'} input + ${cost.output_tokens?.toLocaleString() || '0'} output tokens\n`;
+    }
+    else if (top.input_price !== undefined && top.output_price !== undefined) {
+        result += `Pricing: $${top.input_price}/1M input • $${top.output_price}/1M output\n`;
+    }
+    // Capabilities
+    if (top.capabilities?.context_window || top.context_window) {
+        const contextWindow = top.capabilities?.context_window || top.context_window;
+        result += `Context Window: ${contextWindow.toLocaleString()} tokens\n`;
+    }
+    // Reasoning
+    if (top.reasoning) {
+        result += `\n💡 Why this model?\n${top.reasoning}\n`;
     }
-    if (reasoning) {
-        result += `\n### Why This Model?\n${reasoning}\n`;
+    // Pros and Cons
+    if (top.pros || top.cons) {
+        result += `\n`;
+        if (top.pros && top.pros.length > 0) {
+            result += `✅ Pros:\n`;
+            top.pros.forEach((pro) => {
+                result += `  • ${pro}\n`;
+            });
+        }
+        if (top.cons && top.cons.length > 0) {
+            result += `⚠️  Cons:\n`;
+            top.cons.forEach((con) => {
+                result += `  • ${con}\n`;
+            });
+        }
     }
-    if (alternatives && alternatives.length > 0) {
-        result += `\n### Alternatives\n`;
-        alternatives.forEach((alt, i) => {
-            result += `${i + 1}. **${alt.name}** - ${alt.reason || alt.description || ""}\n`;
+    // Alternative recommendations
+    if (recommendations.length > 1) {
+        result += `\n${"─".repeat(70)}\n`;
+        result += `\nAlternative Options:\n\n`;
+        recommendations.slice(1, 3).forEach((rec, i) => {
+            result += `${i + 2}. ${rec.model_name || rec.name} (${rec.match_score || rec.score || '??'}% match)\n`;
+            result += `   Provider: ${rec.provider}\n`;
+            if (rec.estimated_cost) {
+                result += `   Cost: $${rec.estimated_cost.total_cost_usd?.toFixed(4) || '0.0000'}\n`;
+            }
+            else if (rec.input_price !== undefined) {
+                result += `   Pricing: $${rec.input_price}/1M in • $${rec.output_price}/1M out\n`;
+            }
+            if (rec.reasoning) {
+                result += `   Reason: ${rec.reasoning.substring(0, 150)}${rec.reasoning.length > 150 ? '...' : ''}\n`;
+            }
+            result += `\n`;
         });
     }
-    result += `\n---\n*Powered by [ArchitectGBT](https://architectgbt.com)*`;
+    // Analysis summary
+    if (data.analysis_summary) {
+        result += `${"─".repeat(70)}\n`;
+        result += `\n📊 Analysis Summary:\n${data.analysis_summary}\n\n`;
+    }
+    result += `${"=".repeat(70)}\n`;
+    result += `💎 Powered by ArchitectGBT • https://architectgbt.com`;
     return result;
 }

package/dist/tools/list-models.js

@@ -13,7 +13,7 @@ export const listModelsTool = {
             },
             limit: {
                 type: "number",
-                description: "Maximum number of models to return (default: 10)",
+                description: "Maximum number of models to return (default: 50)",
             },
         },
         required: [],
@@ -23,7 +23,7 @@ const InputSchema = z.object({
     provider: z
         .enum(["OpenAI", "Anthropic", "Google", "Meta", "Mistral", "all"])
         .optional(),
-    limit: z.number().default(10),
+    limit: z.number().default(50),
 });
 export async function handleListModels(args) {
     const input = InputSchema.parse(args);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "architectgbt-mcp",
-  "version": "0.2.1",
+  "version": "0.2.3",
   "description": "Model Context Protocol server for ArchitectGBT - AI architecture recommendations",
   "type": "module",
   "bin": {

package/src/tools/get-recommendation.ts

@@ -126,36 +126,100 @@ export async function handleGetRecommendation(args: unknown) {
 }
 
 function formatRecommendation(data: any): string {
-  const { recommendation, reasoning, alternatives, model } = data;
+  // Handle conversational/off-topic responses
+  if (data.off_topic || data.needs_clarification) {
+    let result = data.message || '';
+    
+    if (data.questions && Array.isArray(data.questions) && data.questions.length > 0) {
+      result += `\n\n📋 To help me recommend the perfect AI model, please tell me:\n`;
+      data.questions.forEach((q: string, i: number) => {
+        result += `${i + 1}. ${q}\n`;
+      });
+    }
+    
+    return result;
+  }
 
-  let result = `## 🎯 AI Model Recommendation\n\n`;
+  const recommendations = data.recommendations || [];
+  
+  if (recommendations.length === 0) {
+    return `❌ No recommendations found. Try describing your project in more detail.`;
+  }
 
-  if (model) {
-    result += `### Recommended: ${model.name}\n`;
-    result += `- **Provider:** ${model.provider}\n`;
-    result += `- **Model ID:** ${model.model_id || "N/A"}\n`;
+  let result = `🎯 AI Model Recommendation — Analysis Complete!\n`;
+  result += `${"=".repeat(70)}\n\n`;
+
+  // Main recommendation
+  const top = recommendations[0];
+  result += `✨ TOP MATCH (${top.match_score || top.score || 95}% match)\n\n`;
+  result += `${top.model_name || top.name}\n`;
+  result += `Provider: ${top.provider}\n`;
+  
+  // Pricing
+  if (top.estimated_cost) {
+    const cost = top.estimated_cost;
+    result += `Estimated Cost: $${cost.total_cost_usd?.toFixed(4) || '0.0000'}\n`;
+    result += `  └─ ${cost.input_tokens?.toLocaleString() || '0'} input + ${cost.output_tokens?.toLocaleString() || '0'} output tokens\n`;
+  } else if (top.input_price !== undefined && top.output_price !== undefined) {
+    result += `Pricing: $${top.input_price}/1M input • $${top.output_price}/1M output\n`;
+  }
 
-    if (model.input_price || model.output_price) {
-      result += `- **Pricing:** $${model.input_price}/1M input, $${model.output_price}/1M output\n`;
-    }
+  // Capabilities
+  if (top.capabilities?.context_window || top.context_window) {
+    const contextWindow = top.capabilities?.context_window || top.context_window;
+    result += `Context Window: ${contextWindow.toLocaleString()} tokens\n`;
+  }
 
-    if (model.context_window) {
-      result += `- **Context Window:** ${model.context_window.toLocaleString()} tokens\n`;
-    }
+  // Reasoning
+  if (top.reasoning) {
+    result += `\n💡 Why this model?\n${top.reasoning}\n`;
   }
 
-  if (reasoning) {
-    result += `\n### Why This Model?\n${reasoning}\n`;
+  // Pros and Cons
+  if (top.pros || top.cons) {
+    result += `\n`;
+    if (top.pros && top.pros.length > 0) {
+      result += `✅ Pros:\n`;
+      top.pros.forEach((pro: string) => {
+        result += `  • ${pro}\n`;
+      });
+    }
+    if (top.cons && top.cons.length > 0) {
+      result += `⚠️  Cons:\n`;
+      top.cons.forEach((con: string) => {
+        result += `  • ${con}\n`;
+      });
+    }
   }
 
-  if (alternatives && alternatives.length > 0) {
-    result += `\n### Alternatives\n`;
-    alternatives.forEach((alt: any, i: number) => {
-      result += `${i + 1}. **${alt.name}** - ${alt.reason || alt.description || ""}\n`;
+  // Alternative recommendations
+  if (recommendations.length > 1) {
+    result += `\n${"─".repeat(70)}\n`;
+    result += `\nAlternative Options:\n\n`;
+    
+    recommendations.slice(1, 3).forEach((rec: any, i: number) => {
+      result += `${i + 2}. ${rec.model_name || rec.name} (${rec.match_score || rec.score || '??'}% match)\n`;
+      result += `   Provider: ${rec.provider}\n`;
+      if (rec.estimated_cost) {
+        result += `   Cost: $${rec.estimated_cost.total_cost_usd?.toFixed(4) || '0.0000'}\n`;
+      } else if (rec.input_price !== undefined) {
+        result += `   Pricing: $${rec.input_price}/1M in • $${rec.output_price}/1M out\n`;
+      }
+      if (rec.reasoning) {
+        result += `   Reason: ${rec.reasoning.substring(0, 150)}${rec.reasoning.length > 150 ? '...' : ''}\n`;
+      }
+      result += `\n`;
     });
   }
 
-  result += `\n---\n*Powered by [ArchitectGBT](https://architectgbt.com)*`;
+  // Analysis summary
+  if (data.analysis_summary) {
+    result += `${"─".repeat(70)}\n`;
+    result += `\n📊 Analysis Summary:\n${data.analysis_summary}\n\n`;
+  }
+
+  result += `${"=".repeat(70)}\n`;
+  result += `💎 Powered by ArchitectGBT • https://architectgbt.com`;
 
   return result;
 }

package/src/tools/list-models.ts

@@ -16,7 +16,7 @@ export const listModelsTool = {
       },
       limit: {
         type: "number",
-        description: "Maximum number of models to return (default: 10)",
+        description: "Maximum number of models to return (default: 50)",
       },
     },
     required: [],
@@ -27,7 +27,7 @@ const InputSchema = z.object({
   provider: z
     .enum(["OpenAI", "Anthropic", "Google", "Meta", "Mistral", "all"])
     .optional(),
-  limit: z.number().default(10),
+  limit: z.number().default(50),
 });
 
 export async function handleListModels(args: unknown) {