Research a Question with a Team of Agents

Add the FutureSearch connector if you haven't already. Then ask Claude:

Compare the major prediction-market platforms: Polymarket, Kalshi, Metaculus, and Manifold Markets, plus any notable peers. For each, cover how it works, its trading volume, and its regulatory status.

Results take about 4 minutes.

Go to futuresearch.ai/app and enter:

Compare the major prediction-market platforms: Polymarket, Kalshi, Metaculus, and Manifold Markets, plus any notable peers. For each, cover how it works, its trading volume, and its regulatory status.

Pass an empty DataFrame as the input: the question is in task. effort_level sets how many agents work the question: low is 3, medium is 4, high is 6.

pip install futuresearch
export FUTURESEARCH_API_KEY=your_key_here  # Get one at futuresearch.ai/app/api-key

import asyncio
import pandas as pd
from futuresearch import create_session
from futuresearch.ops import multi_agent

async def main():
    async with create_session(name="Prediction markets") as session:
        result = await multi_agent(
            session=session,
            task=(
                "Compare the major prediction-market platforms: Polymarket, "
                "Kalshi, Metaculus, and Manifold Markets, plus any notable "
                "peers. For each, cover how it works, its trading volume, "
                "and its regulatory status."
            ),
            input=pd.DataFrame(),
            effort_level="low",
        )
        return result.data

print(asyncio.run(main()).iloc[0]["answer"])

By default the result has a single answer column. To get a multi-field answer, pass a response_schema. See the multi_agent reference.