kazelad/prediction-market-trade-sdk — explained in plain English
Analysis updated 2026-05-18
Build a trading bot that places trades across multiple prediction market venues through one unified API.
Pull market data and prices from platforms like Polymarket, Kalshi, and Metaculus for research or analysis.
Connect an AI assistant such as Claude or Cursor to prediction markets using the built-in MCP support.
Migrate an existing Dome API integration to pmxt using the automatic code conversion tool.
| kazelad/prediction-market-trade-sdk | agutinbaigo28/trading-backtest-kit | alinebm17/finance-api-tool | |
|---|---|---|---|
| Stars | 142 | 142 | 142 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 4/5 | 2/5 |
| Audience | developer | developer | data |
Figures from each repo's GitHub metadata at analysis time.
Requires either a hosted pmxt API key or self-hosted venue credentials to place real trades.
This project, called pmxt, is a software development kit that gives developers one consistent way to read data from and trade on prediction markets. Prediction markets are platforms where people bet on the outcome of future events, such as elections or sports results, and this project connects to many of them, including Polymarket, Kalshi, Limitless, Opinion, Metaculus, and several others, each of which normally has its own different API and data format. Instead of learning each platform's API separately, a developer can use pmxt's single interface to fetch market data across venues, and for a smaller set of supported platforms, place trades as well. There are two main ways to use it. The hosted option means signing up for an API key through the pmxt dashboard, where the service handles account custody, transaction signing, and settling trades on the blockchain on the developer's behalf. The self-hosted option means running the open source server yourself, using your own credentials and private keys, with no dependency on the pmxt hosted service. The project offers software libraries for both Python and TypeScript, plus a command line tool and a plain HTTP interface for any other programming language. It also supports the Model Context Protocol, so AI coding assistants like Claude or Cursor can query and trade on prediction markets directly. For teams migrating from a similar tool called Dome API, pmxt includes an automatic code conversion tool to help move existing code over. Installing it requires Node.js version 18 or higher, or Python 3.8 or higher, depending on which SDK is used. The project is released under the MIT license, and is aimed at developers building trading bots, market research tools, or AI agents that need programmatic access to prediction market data.
pmxt is an SDK giving one consistent API to read data from and trade on multiple prediction market platforms like Polymarket, Kalshi, and Limitless.
Mainly TypeScript. The stack also includes TypeScript, Python, Node.js.
Permissive open-source license, use freely including for commercial purposes, with attribution.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.