Job description...
\",\n \"url\": \"https://linkedin.com/jobs/view/123\"\n }'\n```\n\nResponse:\n```json\n{\n \"success\": true,\n \"message\": \"Job saved to queue\"\n}\n```\n\n#### `GET /api/jobs`\nRetrieve all queued jobs.\n\n```bash\ncurl http://localhost:3002/api/jobs\n```\n\nResponse:\n```json\n{\n \"success\": true,\n \"count\": 5,\n \"data\": [\n {\n \"id\": \"job_1768634456345_0\",\n \"title\": \"Senior Engineer\",\n \"company\": \"TechCorp\",\n \"location\": \"San Francisco, CA\",\n \"workType\": \"Remote\",\n \"description\": \"Job description...
\",\n \"url\": \"https://linkedin.com/jobs/view/123\",\n \"scrapedAt\": \"2026-01-17T04:14:07.702Z\",\n \"source\": \"visual-fallback\"\n }\n ]\n}\n```\n\n#### `DELETE /api/jobs/:id`\nDelete a specific job from the queue.\n\n```bash\ncurl -X DELETE http://localhost:3002/api/jobs/job_1768634456345_0\n```\n\nResponse:\n```json\n{\n \"success\": true,\n \"message\": \"Job deleted\"\n}\n```\n\n#### `DELETE /api/jobs`\nClear all jobs from the queue.\n\n```bash\ncurl -X DELETE http://localhost:3002/api/jobs\n```\n\nResponse:\n```json\n{\n \"success\": true,\n \"message\": \"Job queue cleared\"\n}\n```\n\n## MCP Tools\n\n### Core Tools\n\n| Tool | Description |\n|------|-------------|\n| `saveToFile` | Save text content to a file with auto directory creation |\n| `readFile` | Read file contents as text |\n| `searchFiles` | Search for patterns in files using regex |\n| `executeCommand` | Execute shell commands |\n| `webFetch` | Fetch and parse web content |\n| `scrapeLinkedInJob` | Scrape LinkedIn job postings with Playwright |\n\n### Utility Tools\n\n| Tool | Description |\n|------|-------------|\n| `echoText` | Echo back text (for testing) |\n| `summarizeDirectory` | List directory contents with metadata |\n\n> **📖 For detailed tool documentation, see [TOOL_GUIDE.md](./TOOL_GUIDE.md)**\n\n## Connecting MCP Clients\n\n### Claude Desktop\n\nAdd to your Claude Desktop configuration:\n\n**Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json`\n**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n\n```json\n{\n \"mcpServers\": {\n \"letsmcp\": {\n \"command\": \"node\",\n \"args\": [\"/path/to/letsmcp/dist/index.js\"],\n \"env\": {\n \"PORT\": \"3002\",\n \"HOST\": \"localhost\",\n \"GROQ_API_KEY\": \"your_key_here\"\n }\n }\n }\n}\n```\n\n### Cursor IDE\n\nAdd to Cursor configuration:\n\n**Windows**: `%APPDATA%\\Cursor\\User\\globalStorage\\cursor-mcp\\config.json`\n**macOS**: `~/Library/Application Support/Cursor/User/globalStorage/cursor-mcp/config.json`\n\n```json\n{\n \"mcpServers\": {\n \"letsmcp\": {\n \"command\": \"node\",\n \"args\": [\"/path/to/letsmcp/dist/index.js\"],\n \"env\": {\n \"PORT\": \"3002\",\n \"GROQ_API_KEY\": \"your_key_here\"\n }\n }\n }\n}\n```\n\n> **📖 See [MCP_CONFIG.md](./MCP_CONFIG.md) for more configuration options**\n\n## Integration with External Apps\n\nLetsMCP provides a REST API that can be used by any application. For example, [JobOS](https://github.com/shanebe-ai/JobOS) uses LetsMCP for:\n\n- **Magic Paste**: Extracts job details from pasted text or LinkedIn URLs\n- **Resume Analyst**: AI-powered resume vs job description analysis\n- **Email Drafter**: Generates professional outreach emails\n\n### Example: JobOS Integration\n\n```typescript\n// Check if LetsMCP is available\nconst response = await fetch('http://localhost:3002/api/status');\nconst { hasAI } = await response.json();\n\nif (hasAI) {\n // Use LetsMCP for AI features\n const result = await fetch('http://localhost:3002/api/extract-job', {\n method: 'POST',\n headers: { 'Content-Type': 'application/json' },\n body: JSON.stringify({ url: linkedInUrl })\n });\n}\n```\n\n## Project Structure\n\n```\nletsmcp/\n├── src/\n│ ├── index.ts # Main entry (HTTP server + MCP stdio)\n│ ├── server.ts # MCP server initialization\n│ ├── tools.ts # MCP tool definitions\n│ ├── api/\n│ │ ├── index.ts # API exports\n│ │ └── routes.ts # REST API route handlers\n│ └── ai/\n│ ├── index.ts # AI exports\n│ ├── types.ts # TypeScript interfaces\n│ ├── prompts.ts # Shared AI prompts\n│ ├── service.ts # Unified AI service with fallback\n│ └── providers/\n│ ├── index.ts # Provider exports\n│ ├── groq.ts # Groq (Llama 3.1) provider\n│ ├── claude.ts # Claude (Anthropic) provider\n│ └── gemini.ts # Google Gemini provider\n├── src/__tests__/ # Unit tests\n├── dist/ # Compiled JavaScript\n├── .env.example # Environment template\n├── package.json # Project configuration\n├── tsconfig.json # TypeScript configuration\n└── README.md # This file\n```\n\n## Development\n\n```bash\n# Type checking\nnpm run typecheck\n\n# Run tests\nnpm test\n\n# Run tests in watch mode\nnpm run test:watch\n\n# Build for production\nnpm run build\n```\n\n## Testing\n\n```bash\n# Run all tests\nnpm test\n\n# Test with coverage\nnpm run test:coverage\n\n# Quick health check\ncurl http://localhost:3002/health\n\n# Test AI status\ncurl http://localhost:3002/api/status\n```\n\n## Deployment\n\n### Using PM2 (Recommended for VPS)\n\n```bash\n# Install PM2\nnpm install -g pm2\n\n# Start with PM2\ncd /path/to/letsmcp\nPORT=3002 pm2 start npm --name \"letsmcp\" -- run dev\n\n# Save for auto-restart\npm2 save\npm2 startup\n```\n\n### Cloud Platforms\n\nSee [DEPLOYMENT.md](./DEPLOYMENT.md) for deployment guides for:\n- Railway\n- Render\n- Fly.io\n- Docker\n\n## Troubleshooting\n\n### No AI providers showing\n- Check your `.env` file has valid API keys\n- Restart the server after changing `.env`: `pm2 restart letsmcp --update-env`\n\n### LinkedIn scraping fails\n- LinkedIn may require authentication for some job pages\n- Try using the job description text directly instead of URL\n\n### CORS errors from browser\n- The server has CORS enabled for all origins in development\n- For production, configure allowed origins in `src/index.ts`\n\n### Port already in use\n- Change the port in `.env`: `PORT=3003`\n- Or kill the existing process: `fuser -k 3002/tcp`\n\n## Author\n\n**shanebe-ai** (shanbe@live.com)\n\n## License\n\nMIT\n"}