futuresearch avatar

FutureSearch MCP Server

futuresearch/futuresearch-python

Connects Claude to FutureSearch's research agent API with six core operations: forecast() for predictions, multi_agent() for team research on single questions, agent_map() for running one agent per dataset row, plus rank(), classify(), and dedupe()/merge() for scored research at scale. You'd reach for this when you need Claude to process hundreds or thousands of rows with web research backing each one, like finding regulatory status for 10k drugs or scoring 500 startups. Works in Claude Desktop, Claude Code, and other MCP clients via stdio or HTTP transport. Async variants let you fire off long tasks and retrieve results later if your script crashes. Also integrates with Gemini CLI, Codex, and Cursor as a skill/extension.

tools
10
auth
Secret needed
transport
STDIO
env var
1
last updated
Oct 5, 2026

Tools

10 tools

Public tool metadata for what this MCP can expose to an agent.

futuresearch_forecastForecast questions about the future. Five modes: binary probabilities, numeric percentiles, date percentiles, categorical (one probability per listed outcome), and thresholded (one probability per listed threshold condition).

Forecast questions about the future. Five modes: binary probabilities, numeric percentiles, date percentiles, categorical (one probability per listed outcome), and thresholded (one probability per listed threshold condition).

No parameter schema in public metadata yet.

futuresearch_decisionForecast the outcomes of a decision: the same outcome under each option, e.g. "if we fund this at $0, $300k or $2M, when will it ship?".

Forecast the outcomes of a decision: the same outcome under each option, e.g. "if we fund this at $0, $300k or $2M, when will it ship?".

No parameter schema in public metadata yet.

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

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

No parameter schema in public metadata yet.

futuresearch_agentRun web research agents on each row of a CSV.

Run web research agents on each row of a CSV.

No parameter schema in public metadata yet.

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

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

No parameter schema in public metadata yet.

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

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

No parameter schema in public metadata yet.

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

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

No parameter schema in public metadata yet.

futuresearch_progressCheck progress of a running task.

Check progress of a running task.

No parameter schema in public metadata yet.

futuresearch_resultsRetrieve and save results from a completed task.

Retrieve and save results from a completed task.

No parameter schema in public metadata yet.

futuresearch_cancelCancel a running task.

Cancel a running task.

No parameter schema in public metadata yet.

Sponsors

Configuration

FUTURESEARCH_API_KEY*secret

API key for the FutureSearch service, found at https://futuresearch.ai/app/api-key

Readme

FutureSearch Python SDK

PyPI version License: MIT Python 3.12+

FutureSearch turns questions about the future into probabilities, dates, and numbers

Forecast what will happen, including outcomes of your decisions.

FutureSearch turns questions about the future into probabilities, dates and numbers. Ask about the world ("When will Anthropic IPO?"), or about a decision ("If we give this organization nothing, $250k or $1M, how many lawmakers back its campaign by 2028?") and get predicted outcomes for each option. The decision can be yours or someone else's: a company's, a government's, a public figure's.


View the full README on GitHub

Details

Repository
github.com/futuresearch/futuresearch-python
Website
github.com/futuresearch/futuresearch-python/tree/main/futuresearch-mcp
Package
futuresearch-mcp
Transport
STDIO
Auth
Required
Registry
MCP registry, active