Sieve MCP Server

AI-powered startup due diligence, delivered as an MCP server. Sieve scores startups across 7 IMPACT-X dimensions (Innovators, Market, Product, Advantage, Commerce, Traction, X-Factor) and recommends whether to take a meeting, pass, or request more information.

Quick Start

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "sieve": {
      "command": "uvx",
      "args": ["sieve-mcp"],
      "env": {
        "SIEVE_API_KEY": "your-api-key"
      }
    }
  }
}

Claude Code

claude mcp add sieve -- uvx sieve-mcp

Then set your API key in the environment:

export SIEVE_API_KEY="your-api-key"

Cursor / Windsurf

Add to your MCP settings:

{
  "mcpServers": {
    "sieve": {
      "command": "uvx",
      "args": ["sieve-mcp"],
      "env": {
        "SIEVE_API_KEY": "your-api-key"
      }
    }
  }
}

Environment Variables

Available Tools

Example Usage

A typical workflow in any MCP-compatible AI assistant:

1. Screen a startup:

   "Run a Sieve screen on Acme Corp at acme.com"

2. Poll for progress (analysis takes 2-5 minutes):

   "Check the status of that Sieve analysis"

3. Get results:

   "Show me the Sieve summary for that analysis"

4. Check usage:

   "How many Sieve screens do I have left this month?"

Running with Docker

docker build -t sieve-mcp .
docker run -p 8080:8080 -e SIEVE_API_KEY=your-key sieve-mcp

Development

# Install locally
pip install -e .

# Run in stdio mode (for MCP clients)
sieve-mcp

# Run in HTTP mode (for remote/container deployment)
sieve-mcp http

License

MIT