PredictionToolkit logoPredictionToolkit
Language: English
Search tools
← MCP servers

SDK

pmxt-mcp: bring PMXT research into an assistant

pmxt-mcp exposes PMXT methods as structured tools in an MCP-compatible client. The Node.js wrapper connects to a PMXT HTTP backend for market discovery, book research and configured account or order operations.

3.0/ 5Editorial ratingEditorial assessment
Read pmxt-mcp setup and repositorygithub.com
Available on
MCP
Pricing
Cost follows the backend: hosted calls use the PMXT plan; self-hosting adds your runtime and venue costs.

Screenshots and interface

Product preview1 / 2

MCP setup and backend modes.

What is pmxt-mcp?

The integration gives an assistant a route into an existing PMXT workflow rather than creating a separate dataset. Results inherit PMXT’s venue schemas and adapters. Keep venue and outcome identity visible when interpreting a market or book response.

Choose a hosted PMXT connection or a local pmxt-core runtime before setting credentials. Research, account reads and execution have different access needs. Start with market and book tools, then add only the account operations intended for the workflow and use the MCP client’s action confirmations.

Who it is for

  • Developers integrating prediction-market research with assistants.
  • MCP users with an existing PMXT workflow.
  • Teams defining separate research and account access.

Clients and supported operations

  • Market tools

    Discover supported markets and retrieve structured data.

  • Book research

    Inspect prices and books through PMXT methods.

  • Account reads

    Read supported balances and positions when configured.

  • Optional order tools

    Order methods need a supporting backend, venue and credentials.

  • Hosted or local backend

    Choose PMXT service infrastructure or your own runtime.

  • Inherited venue support

    Schemas and capabilities follow the underlying PMXT adapters.

Development workflows

Connect research tools

Configure the MCP server, start with markets and books, and verify venue and outcome in each response.

Add intended account actions

Configure only required access and keep order operations behind the client’s action-confirmation controls.

Pricing and total cost

Cost follows the backend: hosted calls use the PMXT plan; self-hosting adds your runtime and venue costs.

Installation and configuration

  1. Open the official repository and follow its package and MCP configuration guide.
  2. Choose a backend and supply its environment settings.
  3. Verify market responses identify the intended venue.
  4. Add account or execution access for the intended workflow.

Useful links

What to check

  • Data comes through the configured PMXT backend.
  • Methods inherit venue and backend support.
  • Account and order tools need different access from public research.
  • MCP is an interface, not an independent dataset or exchange.

A closer look

Ratings & reviews

Editorial rating
3.0/ 5
0 user reviews
PredictionToolkit editorial

Our take on pmxt-mcp

The bridge is useful for an existing PMXT workflow. Capabilities and costs follow the backend, while account actions require tighter access than public research.

Assessment criteria
Task usefulness, documented scope, setup clarity, and access and cost transparency.
Assessment date
Documentation reviewed

Documentation-based assessment. How we rate

User reviews 0

The main score switches to user ratings after 5 published reviews.

Be the first to share your experience

A specific example helps the next person choose the right tool.

Related tools

Browse category
PMXT API entry page with an SDK example
PMXT hosted API
Data API3.5Editorial

Search and match markets across venues, retrieve history and use supported trading methods with one hosted key.

  • Access API

Hosted catalog and API plans.

PMXT repository overview and supported venue families
PMXT open-source SDK
SDK3.5Editorial

Use one Python or TypeScript interface across supported prediction venues, with an explicit hosted or local backend choice.

  • Language Python · TypeScript

Local runtime and language clients.

Polygate research and account tools with connection choices
Polygate
SDK3.5Editorial

Run public market research locally or connect a supported web client through a temporary authenticated tunnel.

  • Language Python
  • Access CLI · MCP

A Polymarket-focused MCP server.

Frequently asked questions about pmxt-mcp

Does MCP make every venue tradable?

No. Methods depend on the configured backend and venue.

Can I begin with research only?

Yes. Start with market and book tools without execution credentials.

Is hosted PMXT required?

The documented ecosystem supports hosted and local backends.