{
  "mcpServers": {
    "scout-mcp": {
      "command": "python",
      "args": ["-m", "scout_mcp.mcp_server"],
      "cwd": "/path/to/scout-mcp/src",
      "env": {
        "NEWS_API_KEY": "your-newsapi-key"
      }
    }
  }
}

{
  "scout-mcp": {
    "command": "python",
    "args": ["-m", "scout_mcp.mcp_server"],
    "cwd": "/path/to/scout-mcp/src",
    "env": {
      "NEWS_API_KEY": "your-newsapi-key"
    }
  }
}

{
  "servers": {
    "scout-mcp": {
      "command": "python",
      "args": ["-m", "scout_mcp.mcp_server"],
      "cwd": "/path/to/scout-mcp/src",
      "env": {
        "NEWS_API_KEY": "your-newsapi-key"
      }
    }
  }
}

pip install scout-mcp
scout-mcp  # starts STDIO server for MCP clients

### Running Tests

---

## Docker

// Code block

---

## Tech Stack

| Component | Technology | Purpose |
|-----------|-----------|---------|
| MCP Framework | FastMCP 3.x | Tool registration, STDIO/SSE transport |
| REST API | FastAPI | HTTP endpoints for testing |
| HTTP Client | httpx | Async web scraping |
| HTML Parser | BeautifulSoup4 + lxml | Structured data extraction |
| Search | DuckDuckGo (ddgs) | Free web + news search |
| News | NewsAPI.org | Professional news articles |
| Knowledge | Wikipedia API | Structured background data |
| Validation | Pydantic 2.x | Response model validation |
| Caching | In-memory dict (24h TTL) | Response caching |
| Server | uvicorn | ASGI production server |

---

## Data Sources

| Source | API Key? | Cost | Rate Limit | Reliability |
|--------|----------|------|------------|-------------|
| DuckDuckGo Search | No | Free | Soft limits (backoff) | Variable |
| DuckDuckGo News | No | Free | Soft limits (backoff) | Variable |
| NewsAPI.org | Yes | Free tier | 100 req/day | High |
| Wikipedia API | No | Free | Unlimited (polite) | Very High |
| Web Scraping (httpx) | No | Free | Per-site limits | Medium |
| Social Profile Detection | No | Free | Via DuckDuckGo | Variable |

### Future Sources (Planned)

- Crunchbase API (funding data)
- SimilarWeb API (traffic data)
- GitHub API (developer tools)
- SEMrush API (SEO data)

---

## Contributing

1. Fork the repository
2. Create a feature branch: `git checkout -b feature/my-feature`
3. Install dev dependencies: `pip install -e ".[dev]"`
4. Make your changes
5. Run tests: `pytest`
6. Submit a pull request

### Adding a New Data Source

1. Create `src/scout_mcp/sources/your_source.py`
2. Implement async functions that return structured data
3. Add the source to relevant tools in `src/scout_mcp/tools/`
4. Add per-source confidence scoring
5. Update this README

### Adding a New Tool

1. Create `src/scout_mcp/tools/your_tool.py`
2. Add a Pydantic model in `models.py` (include `confidence_breakdown` and `data_quality_grade`)
3. Register in `mcp_server.py` with `@mcp.tool()`
4. Add REST endpoint in `server.py`
5. Update this README

---

## License

MIT

---

  Built with FastMCP, httpx, BeautifulSoup4, Pydantic
  Scout MCP v0.1.0

---

  🚀 Also check out  — cross-chain DeFi intelligence for AI agents. Risk-scored yields, Monte Carlo simulations, whale tracking across 86 chains.