# Quick Start Guide
Get your Personal Productivity MCP Server running in 5 minutes.
## Prerequisites
- Docker and Docker Compose installed
- Claude Desktop app (for MCP integration)
## Step 1: Environment Setup
The `.env` file has already been created from the template. Review it if you need to customize any settings:
```bash
cat .env
```
Default settings should work fine for getting started.
## Step 2: Build and Start the Server
### Option A: Using the Quick Start Script
```bash
./scripts/start.sh
```
### Option B: Manual Docker Commands
```bash
# Build the container
docker-compose build
# Start the server
docker-compose up -d
# View logs
docker-compose logs -f
```
## Step 3: Verify the Server is Running
Check that the container is running:
```bash
docker ps | grep personal-productivity-mcp
```
Check the logs for any errors:
```bash
docker-compose logs
```
You should see output like:
```
Starting Personal Productivity MCP Server
Server version: 0.1.0
Database initialized
Server running on stdio
```
## Step 4: Connect to Claude Desktop
### macOS Configuration
Edit your Claude Desktop config file:
```bash
code ~/Library/Application\ Support/Claude/claude_desktop_config.json
```
Add this configuration:
```json
{
"mcpServers": {
"personal-productivity": {
"command": "docker",
"args": [
"exec",
"-i",
"personal-productivity-mcp",
"python",
"-m",
"mcp_server.server"
]
}
}
}
```
### Restart Claude Desktop
Completely quit and restart Claude Desktop for the changes to take effect.
## Step 5: Test the Server
In Claude Desktop, try these commands:
1. **Search for jobs:**
```
Search for software engineering jobs at Anthropic
```
2. **Track an application:**
```
Track a new job application:
- Company: Anthropic
- Title: Research Engineer
- URL: https://jobs.lever.co/anthropic/example
- Status: interested
- Notes: Found through Greenhouse API
```
3. **View your pipeline:**
```
Show me my application pipeline statistics
```
## Troubleshooting
### Container won't start
Check Docker logs:
```bash
docker-compose logs
```
Common issues:
- Port 3000 already in use (not applicable for stdio mode)
- Database permissions (ensure `./data` directory exists and is writable)
### Claude Desktop can't connect
1. Make sure the container is running: `docker ps`
2. Verify the config path is correct for your OS
3. Check Claude Desktop logs (Help > Show Logs)
4. Restart Claude Desktop completely
### Scrapers return no results
- Check your internet connection
- Try a different company name (e.g., "anthropic", "netflix")
- Check logs: `docker-compose logs | grep ERROR`
### Database errors
Reset the database:
```bash
docker-compose down
rm data/productivity.db
docker-compose up -d
```
## Development Mode
### Run tests
```bash
./scripts/test.sh
```
Or manually:
```bash
docker-compose run --rm mcp-server pytest -v
```
### View database contents
```bash
docker-compose exec mcp-server sqlite3 /app/data/productivity.db
```
Example queries:
```sql
-- View all applications
SELECT * FROM applications;
-- Count by status
SELECT status, COUNT(*) FROM applications GROUP BY status;
```
### Access container shell
```bash
docker-compose exec mcp-server /bin/bash
```
## Next Steps
1. **Customize scrapers**: Edit `src/mcp_server/domains/career/scrapers/` to add more job boards
2. **Add target companies**: Use the tools to track companies you're interested in
3. **Build automation**: Set up n8n or other tools to call the MCP server on a schedule
4. **Expand domains**: Add fitness or family domains when ready (see brief.md)
## Useful Commands
```bash
# Stop the server
docker-compose down
# Restart the server
docker-compose restart
# View logs in real-time
docker-compose logs -f
# Rebuild after code changes
docker-compose up --build -d
# Clean everything (including database)
docker-compose down -v
rm -rf data/*
```
## Getting Help
- Check the main README.md for architecture details
- Review brief.md for the full project plan
- Look at test files in `tests/` for usage examples
- Check server logs for error messages
## Phase 2 Preview
Coming soon:
- Generic career page scraper for companies without Greenhouse/Lever
- Job details extraction with full descriptions
- Resume analysis and fit scoring
- Cover letter point generation
Stay tuned!
