FutureSearch MCP Server

{
  "mcpServers": {
    "futuresearch-mcp-server": {
      "command": "uvx",
      "args": [
        "futuresearch-mcp"
      ],
      "env": {
        "FUTURESEARCH_API_KEY": "your-value"
      }
    }
  }
}

claude mcp add --env FUTURESEARCH_API_KEY=your-value --transport stdio futuresearch-mcp-server -- uvx futuresearch-mcp

{
  "mcpServers": {
    "futuresearch-mcp-server": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "futuresearch-mcp"
      ],
      "env": {
        "FUTURESEARCH_API_KEY": "your-value"
      }
    }
  }
}

{
  "inputs": [
    {
      "type": "promptString",
      "id": "futuresearch_api_key",
      "description": "FUTURESEARCH_API_KEY",
      "password": true
    }
  ],
  "servers": {
    "futuresearch-mcp-server": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "futuresearch-mcp"
      ],
      "env": {
        "FUTURESEARCH_API_KEY": "${input:futuresearch_api_key}"
      }
    }
  }
}

Config formats checked against the official docs on Oct 7, 2026: modelcontextprotocol.io (opens in a new tab), code.claude.com (opens in a new tab), cursor.com (opens in a new tab), code.visualstudio.com (opens in a new tab).

• futuresearch_forecast

  Forecast questions about the future. Five modes: binary probabilities, numeric

• futuresearch_decision

  Forecast the outcomes of a decision: the same outcome under each option, e.g.

• futuresearch_multi_agent

  Deep parallel research: deploy a team of direction agents per row exploring different angles, then synthesize.

• futuresearch_agent

  Run web research agents on each row of a CSV.

• futuresearch_browse_lists

  Browse available reference lists of well-known entities (S&P 500, FTSE 100, countries, universities, etc.).

• futuresearch_use_list

  Import a reference list into your session and make it available via artifact_id for other futuresearch tools.

• futuresearch_upload_data

  Upload data from a URL or local file. Returns an artifact_id for use in processing tools.

• futuresearch_progress

  Check progress of a running task.

• futuresearch_results

  Retrieve and save results from a completed task.

• futuresearch_cancel

  Cancel a running task.

Most users don't need to run the MCP server locally. Use the hosted remote server at — it authenticates via OAuth, no API key needed. See the setup guide. The instructions below are for self-hosted or advanced use cases where an API key is required.

• claude-code-plugin
• forecasting
• mcp
• multi-agent
• prediction
• sdk
• decision-making
• scenario-analysis