polydev-ai 1.4.0 → 1.4.2

package/README.md

@@ -1,359 +1,160 @@
-# Polydev AI Website
+# Polydev AI MCP Server
 
-Advanced Model Context Protocol platform with comprehensive multi-LLM integrations, subscription-based CLI access, OAuth bridges, and advanced tooling for AI development.
+Get diverse AI perspectives from multiple LLMs via Model Context Protocol (MCP). Supports Cline, Claude Code, and other MCP clients with local CLI detection and remote AI perspectives.
 
 ## Features
 
-### 🤖 **Comprehensive LLM Integration**
-- **API-Based Providers**: Direct integration with 8+ providers (Anthropic, OpenAI, Google, etc.)
-- **Subscription-Based CLI Access**: Use your existing ChatGPT Plus, Claude Pro, GitHub Copilot subscriptions
-- **Unified Interface**: Single API for all providers with consistent streaming responses
-- **Auto-Detection**: Automatic CLI tool discovery and path configuration
-
-### 🔧 **CLI Provider Support**
-- **Codex CLI**: Access GPT-5 with high reasoning through ChatGPT subscription
-- **Claude Code CLI**: Use Claude via Anthropic subscription
-- **Gemini CLI**: Google Cloud authentication integration
-- **GitHub Copilot**: VS Code Language Model API integration
-
-### 🛠 **Advanced Tooling**
-- **Model Context Protocol (MCP)**: Hosted MCP server with OAuth authentication (like Vercel)
-- **Multi-Authentication**: Both OAuth and API token support for maximum flexibility
-- **Process Execution**: Cross-platform CLI management with timeout handling
-- **Path Auto-Discovery**: Smart detection of CLI installations across Windows, macOS, Linux
-- **Real-time Status**: Live CLI availability and authentication checking
-
-### 🔒 **Security & Authentication**
-- **Encrypted Storage**: Browser-based API key encryption
-- **OAuth Bridges**: Secure authentication flows
-- **Subscription Auth**: No API costs - use existing subscriptions
-- **Local Storage**: Keys never leave your device
-
-### 📊 **Monitoring & Analytics**  
-- **PostHog Integration**: Advanced user analytics and feature tracking
-- **BetterStack Monitoring**: System health and performance monitoring
-- **Upstash Redis**: High-performance caching layer
-- **Supabase Auth**: Robust authentication system
-
-## Tech Stack
-
-### **Frontend**
-- **Framework**: Next.js 15 with App Router
-- **UI Library**: React 18 with TypeScript
-- **Styling**: Tailwind CSS with shadcn/ui components
-- **Icons**: Lucide React
-- **State Management**: React hooks with custom providers
-
-### **LLM Integration** 
-- **API Handlers**: Custom TypeScript handlers for each provider
-- **CLI Integration**: Cross-platform process execution utilities
-- **Streaming**: Server-Sent Events for real-time responses
-- **Authentication**: Both API key and subscription-based authentication
-
-### **Backend Services**
-- **Analytics**: PostHog for user tracking and feature analytics
-- **Monitoring**: BetterStack for system health and logging
-- **Caching**: Upstash Redis for high-performance data caching
-- **Authentication**: Supabase for user management and auth flows
-- **Database**: Supabase PostgreSQL for user data
-
-### **Security & Storage**
-- **Encryption**: Browser SubtleCrypto API for client-side encryption
-- **Storage**: Local browser storage with encrypted API keys
-- **CORS**: Configured for secure cross-origin requests
-
-### **Development & Deployment**
-- **Package Manager**: npm with Node.js 18+
-- **Build System**: Next.js with TypeScript compilation
-- **Deployment**: Vercel with automatic deployments
-- **Environment**: Support for multiple deployment environments
-
-## Getting Started
-
-### **Prerequisites**
-- Node.js 18+ 
-- npm or yarn package manager
-- (Optional) CLI tools for subscription-based access:
-  - Codex CLI for ChatGPT Plus integration
-  - Claude Code CLI for Anthropic Pro integration
-  - Gemini CLI for Google Cloud integration
-  - VS Code with GitHub Copilot for Copilot integration
-
-### **Installation**
-
-1. **Clone the repository**
-```bash
-git clone <repository-url>
-cd polydev-website
-```
+- 🤖 **Multi-Model AI Perspectives**: Get responses from GPT-4, Claude, Gemini, and more
+- 🔧 **Local CLI Detection**: Automatically detect and use Claude Code, Codex CLI, Gemini CLI
+- ⚡ **Smart Caching**: Intelligent refresh intervals based on CLI status
+- 🔄 **Fallback Support**: Local CLI + remote perspectives for comprehensive responses
+- 🔐 **Secure**: Token-based authentication with your Polydev account
 
-2. **Install dependencies**
-```bash
-npm install
-```
+## Installation
 
-3. **Set up environment variables** (see Environment Variables section)
+### For Claude Code Users
 
-4. **Start development server**
 ```bash
-npm run dev
-```
+# Install globally
+npm install -g polydev-ai
 
-5. **Open the application**
-Navigate to [http://localhost:3000](http://localhost:3000) to view the application.
+# Or use directly with npx
+npx polydev-ai
+```
 
-### **Quick Configuration**
+### For Cline Users
+
+Add to your MCP settings (`.cline/mcp_servers.json`):
+
+```json
+{
+  "mcpServers": {
+    "polydev": {
+      "command": "npx",
+      "args": ["polydev-ai"],
+      "env": {
+        "POLYDEV_USER_TOKEN": "your_token_here"
+      }
+    }
+  }
+}
+```
 
-1. **API Key Setup**: Go to Settings → API Keys tab to configure traditional API access
-2. **CLI Setup**: Go to Settings → CLI Subscriptions tab to set up subscription-based access
-3. **Provider Selection**: Choose your preferred LLM provider from the dropdown
-4. **Test Integration**: Use the chat interface to test your configuration
+## Configuration
 
-## Environment Variables
+### Get Your Token
 
-Create a `.env.local` file with the following variables:
+1. Sign up at [polydev.ai](https://polydev.ai)
+2. Go to your dashboard and copy your MCP token
+3. Set the environment variable:
 
-```env
-# Supabase
-NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
-NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
-SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key
+```bash
+export POLYDEV_USER_TOKEN="pd_your_token_here"
+```
 
-# PostHog Analytics
-NEXT_PUBLIC_POSTHOG_KEY=your_posthog_key
-NEXT_PUBLIC_POSTHOG_HOST=https://us.i.posthog.com
+### Claude Code Integration
 
-# Upstash Redis
-UPSTASH_REDIS_REST_URL=your_upstash_redis_url
-UPSTASH_REDIS_REST_TOKEN=your_upstash_redis_token
+Add to your Claude Code MCP configuration:
 
-# BetterStack Logging
-BETTERSTACK_LOGS_TOKEN=your_betterstack_token
+```json
+ "polydev": {
+          "disabled": false,
+          "timeout": 120,
+          "type": "http",
+          "url": "https://www.polydev.ai/api/mcp"
+        },
 ```
 
-## CLI Provider Setup
-
-### **Codex CLI (ChatGPT Plus Integration)**
-
-1. **Install Codex CLI**:
-   - Download from [OpenAI's official repository](https://openai.com/chatgpt/desktop)
-   - Ensure you have an active ChatGPT Plus subscription
-
-2. **Authentication**:
-   ```bash
-   codex auth
-   ```
-
-3. **Verify Installation**:
-   ```bash
-   codex --version
-   ```
-
-### **Claude Code CLI (Anthropic Pro Integration)**
-
-1. **Install Claude Code CLI**:
-   - Follow instructions at [Claude Code Documentation](https://docs.anthropic.com/en/docs/claude-code)
-   - Requires active Claude Pro subscription
-
-2. **Authentication**:
-   ```bash
-   claude login
-   ```
-
-3. **Verify Installation**:
-   ```bash
-   claude --version
-   ```
-
-### **Gemini CLI (Google Cloud Integration)**
-
-1. **Install Google Cloud CLI**:
-   ```bash
-   # macOS
-   brew install google-cloud-sdk
-   
-   # Windows
-   # Download from https://cloud.google.com/sdk/docs/install
-   
-   # Linux
-   curl https://sdk.cloud.google.com | bash
-   ```
-
-2. **Authentication**:
-   ```bash
-   gcloud auth login
-   gcloud auth application-default login
-   ```
-
-### **GitHub Copilot Integration**
-
-1. **Install VS Code** with GitHub Copilot extension
-2. **Authentication**: Sign in with your GitHub account that has Copilot access
-3. **Verification**: The application will detect VS Code and Copilot availability automatically
-
-## API Provider Configuration
-
-### **Setting Up API Keys**
-
-1. Navigate to Settings → API Keys tab
-2. Select your preferred provider from the dropdown
-3. Enter your API key (encrypted automatically)
-4. Choose your preferred model
-5. Test the configuration
-
-### **Supported API Providers**
-
-| Provider | Models | Context Window | Features |
-|----------|--------|---------------|----------|
-| **Anthropic** | Claude 3.5 Sonnet, Haiku, Opus | 200K tokens | Best for reasoning and code |
-| **OpenAI** | GPT-4o, GPT-4 Turbo, GPT-3.5 | 128K tokens | Versatile, widely adopted |
-| **Google Gemini** | Gemini 1.5 Pro, Flash | 1M+ tokens | Large context window |
-| **OpenRouter** | 100+ models | Varies | Access to multiple providers |
-| **Groq** | Open-source models | Varies | Ultra-fast inference |
-| **Perplexity** | Search-optimized models | Varies | AI search and reasoning |
-| **DeepSeek** | Reasoning models | Varies | Advanced reasoning capabilities |
-| **Mistral AI** | European AI models | Varies | Strong performance, EU-based |
+## Available Tools
+
+- `get_perspectives` - Get AI responses from multiple models
+- `force_cli_detection` - Force detection of local CLI tools
+- `get_cli_status` - Check status of CLI tools (Claude Code, Codex, Gemini)
+- `send_cli_prompt` - Send prompts to local CLI with perspectives fallback
 
 ## Usage Examples
 
-### **Basic Chat Interface**
+### Get Multi-Model Perspectives
 
-1. Configure your preferred provider (API key or CLI)
-2. Select a model from the dropdown
-3. Start chatting in the main interface
-4. Switch providers anytime without losing conversation history
+```typescript
+// Use the get_perspectives tool
+{
+  "prompt": "How do I optimize React performance?",
+  "models": ["gpt-4", "claude-3-sonnet", "gemini-pro"]
+}
+```
 
-### **CLI Provider Usage**
+### Check CLI Status
 
 ```typescript
-// The application automatically detects CLI availability
-// Users can configure custom paths if needed
-
-// Example: Using Codex CLI for high reasoning
-const response = await llmService.createCliMessage(
-  'codex',
-  'You are a helpful AI assistant',
-  [{ role: 'user', content: 'Explain quantum computing' }],
-  { temperature: 0.7 }
-)
+// Check all CLI tools
+await get_cli_status({})
+
+// Check specific tool
+await get_cli_status({ provider_id: "claude_code" })
 ```
 
-### **API Provider Usage**
+### Send CLI Prompt with Fallback
 
 ```typescript
-// Standard API key usage
-const response = await llmService.createMessage(
-  'You are a helpful AI assistant',
-  [{ role: 'user', content: 'Write a Python function' }],
-  { provider: 'anthropic', model: 'claude-3-5-sonnet-20241022' }
-)
+await send_cli_prompt({
+  provider_id: "claude_code",
+  prompt: "Write a Python function to parse JSON",
+  mode: "args"
+})
 ```
 
-## Architecture Overview
+## Supported CLI Tools
 
-### **CLI Integration Architecture**
+- **Claude Code** (`claude`) - Anthropic's official CLI
+- **Codex CLI** (`codex`) - OpenAI's code CLI
+- **Gemini CLI** (`gemini`) - Google's Gemini CLI
 
-```
-┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
-│   Frontend UI   │────│  Process Utils   │────│   CLI Tools     │
-│   (React/TS)    │    │  (Node.js)       │    │   (External)    │
-└─────────────────┘    └──────────────────┘    └─────────────────┘
-         │                        │                        │
-         │                        │                        │
-         ▼                        ▼                        ▼
-┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
-│ LLM Service     │    │ CLI Handlers     │    │ Subscriptions   │
-│ (Unified API)   │    │ (Per Provider)   │    │ (ChatGPT+, etc) │
-└─────────────────┘    └──────────────────┘    └─────────────────┘
-```
+## Smart Refresh System
+
+The MCP server automatically detects CLI status changes:
 
-### **Security Model**
+- **Unavailable CLIs**: Check every 2 minutes
+- **Unauthenticated CLIs**: Check every 3 minutes  
+- **Working CLIs**: Check every 10 minutes
+- **Fallback detection**: Check every 5 minutes
+
+## Environment Variables
 
-- **API Keys**: Encrypted using browser SubtleCrypto API
-- **Local Storage**: Keys never leave your device
-- **CLI Authentication**: Uses existing subscription authentication
-- **Process Isolation**: CLI processes run in isolated environments
+- `POLYDEV_USER_TOKEN` - Your Polydev authentication token (required)
+- `POLYDEV_CLI_DEBUG` - Enable CLI debugging output
+- `CLAUDE_CODE_PATH` - Custom path to Claude Code CLI
+- `CODEX_CLI_PATH` - Custom path to Codex CLI
+- `GEMINI_CLI_PATH` - Custom path to Gemini CLI
 
 ## Troubleshooting
 
-### **CLI Provider Issues**
-
-**Codex CLI not detected:**
-1. Verify ChatGPT Plus subscription is active
-2. Check installation path: `which codex`
-3. Re-run authentication: `codex auth`
-4. Configure custom path in Settings → CLI Subscriptions
-
-**Claude Code CLI authentication failed:**
-1. Ensure Claude Pro subscription
-2. Run `claude login` manually
-3. Check network connectivity
-4. Verify CLI version compatibility
-
-**Gemini CLI setup issues:**
-1. Install Google Cloud SDK completely
-2. Run both `gcloud auth` commands
-3. Enable required APIs in Google Cloud Console
-4. Check quota limits
-
-**GitHub Copilot not available:**
-1. Install VS Code with Copilot extension
-2. Sign in to GitHub account with Copilot access
-3. Restart the application
-4. Check VS Code Language Model API availability
-
-### **API Provider Issues**
-
-**API key validation failed:**
-1. Verify the key is correctly copied (no extra spaces)
-2. Check if the key has required permissions
-3. Ensure sufficient account credits/quota
-4. Try regenerating the API key
-
-**Connection timeout:**
-1. Check internet connectivity
-2. Verify firewall settings
-3. Try different model/provider
-4. Increase timeout in settings
-
-**Model not available:**
-1. Check provider documentation for model availability
-2. Verify your account tier supports the model
-3. Try alternative models from the same provider
-
-## Development
-
-### **Adding New CLI Providers**
-
-1. **Create handler** in `src/lib/llm/handlers/`
-2. **Update types** in `src/types/api-configuration.ts`
-3. **Add configuration** to `CLI_PROVIDERS` constant
-4. **Update UI** in `CliProviderConfiguration.tsx`
-5. **Register handler** in `LLMService`
-
-### **Adding New API Providers**
-
-1. **Create handler** implementing `ApiHandler` interface
-2. **Update `ApiProvider` type** and `PROVIDERS` configuration
-3. **Add API key fields** to `ApiConfiguration`
-4. **Update UI** components for new provider
-5. **Add authentication logic**
-
-## Performance Optimization
-
-### **CLI Response Optimization**
-- CLI processes are cached and reused when possible
-- Streaming responses reduce perceived latency
-- Process timeouts prevent hanging connections
-- Cross-platform path detection minimizes setup time
-
-### **API Response Optimization**
-- Server-Sent Events for real-time streaming
-- Connection pooling for API requests
-- Response caching for repeated queries
-- Automatic retry logic with exponential backoff
-
-## Health Check
-
-The application includes a health check endpoint at `/api/health` for monitoring purposes.
-# Trigger deployment
+### CLI Not Detected
+
+1. Ensure the CLI is installed and in your PATH
+2. Check authentication: `claude auth status` / `codex login status`
+3. Enable debugging: `export POLYDEV_CLI_DEBUG=1`
+
+### Token Issues
+
+1. Verify your token at [polydev.ai/dashboard](https://polydev.ai/dashboard)
+2. Ensure the token starts with `pd_`
+3. Check environment variable is set correctly
+
+### Permission Errors
+
+```bash
+# Fix npm permissions
+npm config set prefix '~/.npm-global'
+export PATH=~/.npm-global/bin:$PATH
+```
+
+## Support
+
+- 📧 Email: [support@polydev.ai](mailto:support@polydev.ai)
+- 📖 Docs: [polydev.ai/docs](https://polydev.ai/docs)
+- 🐛 Issues: [GitHub Issues](https://github.com/backspacevenkat/polydev-website/issues)
+
+## License
+
+MIT License - see LICENSE file for details.

package/package.json

@@ -1,102 +1,47 @@
 {
   "name": "polydev-ai",
-  "version": "1.4.0",
-  "description": "Agentic workflow assistant with CLI integration - get diverse perspectives from multiple LLMs when stuck or need enhanced reasoning",
+  "version": "1.4.2",
+  "description": "Get diverse AI perspectives from multiple LLMs via MCP - supports Cline, Claude Code, and other MCP clients",
+  "main": "stdio-wrapper.js",
+  "bin": {
+    "polydev-ai": "./stdio-wrapper.js"
+  },
+  "scripts": {
+    "start": "node server.js",
+    "test": "node server.js --help"
+  },
   "keywords": [
     "mcp",
     "model-context-protocol",
     "ai",
+    "llm",
+    "perspectives",
+    "multi-model",
+    "cline",
     "claude",
     "openai",
-    "perspectives",
-    "cli",
-    "agent",
-    "assistant"
+    "anthropic",
+    "gemini",
+    "polydev"
   ],
-  "author": "Polydev AI",
+  "author": "Polydev AI <support@polydev.ai>",
   "license": "MIT",
-  "private": false,
-  "main": "mcp/server.js",
-  "bin": {
-    "polydev-stdio": "mcp/stdio-wrapper.js"
+  "engines": {
+    "node": ">=16.0.0"
   },
   "files": [
-    "mcp/",
-    "lib/",
-    "README.md",
-    "LICENSE"
+    "stdio-wrapper.js",
+    "cliManager.js",
+    "manifest.json",
+    "README.md"
   ],
-  "scripts": {
-    "dev": "next dev",
-    "build": "next build",
-    "start": "next start",
-    "lint": "next lint",
-    "test": "jest",
-    "test:watch": "jest --watch",
-    "test:coverage": "jest --coverage",
-    "test:api": "jest __tests__/api",
-    "test:vm": "jest __tests__/api/vm",
-    "test:admin": "jest __tests__/api/admin",
-    "mcp": "node mcp/server.js",
-    "mcp-stdio": "node mcp/stdio-wrapper.js",
-    "cli-detect": "node -e \"const CLIManager = require('./lib/cliManager').default; const m = new CLIManager(); m.forceCliDetection().then(console.log);\""
-  },
   "repository": {
     "type": "git",
-    "url": "https://github.com/polydev-ai/perspectives-mcp.git"
+    "url": "https://github.com/backspacevenkat/polydev-website.git",
+    "directory": "mcp"
   },
-  "homepage": "https://polydev.ai",
+  "homepage": "https://polydev.ai/docs/mcp-integration",
   "bugs": {
-    "url": "https://github.com/polydev-ai/perspectives-mcp/issues"
-  },
-  "dependencies": {
-    "@hello-pangea/dnd": "^18.0.1",
-    "@radix-ui/react-dialog": "^1.1.15",
-    "@radix-ui/react-progress": "^1.1.7",
-    "@radix-ui/react-select": "^2.2.6",
-    "@radix-ui/react-slot": "^1.2.3",
-    "@radix-ui/react-tabs": "^1.1.13",
-    "@supabase/ssr": "^0.4.0",
-    "@supabase/supabase-js": "^2.45.0",
-    "@upstash/redis": "^1.34.0",
-    "class-variance-authority": "^0.7.1",
-    "clsx": "^2.1.1",
-    "date-fns": "^4.1.0",
-    "dotenv": "^17.2.2",
-    "framer-motion": "^12.23.22",
-    "lucide-react": "^0.542.0",
-    "marked": "^16.2.1",
-    "next": "15.0.0",
-    "polydev-ai": "^1.2.0",
-    "posthog-js": "^1.157.2",
-    "prismjs": "^1.30.0",
-    "react": "^18.3.1",
-    "react-dom": "^18.3.1",
-    "resend": "^6.0.2",
-    "shelljs": "^0.8.5",
-    "sonner": "^2.0.7",
-    "stripe": "^18.5.0",
-    "tailwind-merge": "^3.3.1",
-    "tailwindcss-animate": "^1.0.7",
-    "ts-node": "^10.9.2",
-    "use-debounce": "^10.0.6",
-    "which": "^5.0.0"
-  },
-  "devDependencies": {
-    "@testing-library/jest-dom": "^6.9.1",
-    "@testing-library/react": "^16.3.0",
-    "@types/jest": "^30.0.0",
-    "@types/node": "^20.16.1",
-    "@types/react": "^18.3.3",
-    "@types/react-dom": "^18.3.0",
-    "autoprefixer": "^10.4.20",
-    "eslint": "^8.57.0",
-    "eslint-config-next": "15.0.0",
-    "jest": "^30.2.0",
-    "jest-environment-jsdom": "^30.2.0",
-    "postcss": "^8.4.41",
-    "tailwindcss": "^3.4.10",
-    "ts-jest": "^29.4.4",
-    "typescript": "^5.5.4"
+    "url": "https://github.com/backspacevenkat/polydev-website/issues"
   }
-}
+}

package/{mcp/stdio-wrapper.js → stdio-wrapper.js}

@@ -4,7 +4,7 @@
 const fs = require('fs');
 const path = require('path');
 const os = require('os');
-const { CLIManager } = require('../lib/cliManager');
+const { CLIManager } = require('./cliManager');
 
 function ensureWritableTmpDir() {
   const candidates = [
@@ -505,9 +505,13 @@ class StdioMCPWrapper {
           raw: true // Flag to indicate this is pre-formatted content
         };
       } else if (remoteResponse.error) {
+        // Normalize error - handle both string and object formats
+        const errorMessage = typeof remoteResponse.error === 'string'
+          ? remoteResponse.error
+          : (remoteResponse.error?.message || JSON.stringify(remoteResponse.error) || 'Remote perspectives failed');
         return {
           success: false,
-          error: remoteResponse.error.message || 'Remote perspectives failed',
+          error: errorMessage,
           timestamp: new Date().toISOString()
         };
       } else {
@@ -564,8 +568,9 @@ class StdioMCPWrapper {
     } else {
       // Both failed
       combinedResult.success = false;
-      const cliErrors = localResults.map(cli => `${cli.provider_id}: ${cli.error}`).join('; ');
-      combinedResult.error = `All local CLIs failed: ${cliErrors}; Perspectives also failed: ${perspectivesResult.error}`;
+      const cliErrors = localResults.map(cli => `${cli.provider_id}: ${cli.error || 'Unknown error'}`).join('; ');
+      const perspectivesError = perspectivesResult?.error || 'Unknown remote error';
+      combinedResult.error = `All local CLIs failed: ${cliErrors}; Perspectives also failed: ${perspectivesError}`;
     }
 
     return combinedResult;