# DealLens MCP

Real estate investment deal analysis for Claude Desktop and MCP-compatible AI clients. Analyze rentals, compare deals, stress-test scenarios, and evaluate BRRRR/flip strategies — all using free government data.

## Quick Start

### 1. Get Free API Keys (5 minutes)

- **HUD API Token:** https://www.huduser.gov/portal/dataset/fmr-api.html
- **FRED API Key:** https://fred.stlouisfed.org/docs/api/api_key.html
- **Census API Key:** https://api.census.gov/data/key_signup.html

### 2. Install

```bash
npm install -g deallens-mcp
```

### 3. Configure Claude Desktop

Add to your `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "deallens": {
      "command": "npx",
      "args": ["deallens-mcp"],
      "env": {
        "HUD_API_TOKEN": "your-hud-token",
        "FRED_API_KEY": "your-fred-key",
        "CENSUS_API_KEY": "your-census-key"
      }
    }
  }
}
```

Restart Claude Desktop. You should see DealLens tools available.

## Tools

### `analyze_rental_deal`

Full investment analysis: mortgage, expenses, cash flow, cap rate, cash-on-cash, DSCR, 1% rule, and a verdict.

> "Analyze this rental: $285K, 3bed/2bath in ZIP 77019, expected rent $2,100/month. I'm putting 20% down."

### `get_area_rents`

HUD Fair Market Rents for any US ZIP code.

> "What are the fair market rents in ZIP 30301?"

### `get_mortgage_rates`

Current 30-year and 15-year fixed rates from the Federal Reserve.

> "What are current mortgage rates?"

### `compare_deals`

Side-by-side comparison of 2-5 properties, ranked by cash-on-cash return.

> "Compare: Property A $200K rent $1,500 in 30301 vs Property B $350K rent $2,800 in 77019. Both 20% down."

### `sensitivity_analysis`

Stress-test a deal: what if rates go up? What if vacancy hits 10%?

> "What happens to that deal if interest rates go from 6% to 8%?"

### `estimate_expenses`

Estimate monthly operating expenses using industry-standard ratios.

> "Estimate expenses for a $2,000/month rental, SFH built in 1985."

### `brrrr_analysis`

Buy-Rehab-Rent-Refinance-Repeat strategy calculator.

> "BRRRR analysis: $150K purchase, $40K rehab, ARV $250K, expected rent $1,800/month in ZIP 77019."

### `flip_analysis`

Fix-and-flip profitability analysis.

> "Flip analysis: buying at $120K, $30K rehab, ARV $200K, 4 month hold, hard money financing."

## Data Sources

All data comes from free US government APIs:

| Source | What It Provides |
|--------|-----------------|
| HUD User | Fair Market Rents by ZIP code |
| FRED (Federal Reserve) | Current mortgage rates |
| Census Bureau | Median home values, income, rent by area |
| BLS | Employment data (v2) |

## Development

```bash
git clone https://github.com/yourusername/deallens-mcp.git
cd deallens-mcp
npm install
npm run build
npm test
```

### Testing with MCP Inspector

```bash
npx @modelcontextprotocol/inspector build/index.js
```

## License

MIT