Skip to content
MCP server

Oracle3 prediction-market arbitrage checks

Kalshi and Polymarket data, fee-aware arbitrage checks and paper trading for AI agents

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
1.3.0
Listing updated
1 day ago

About Oracle3 prediction-market arbitrage checks

Oracle3 prediction-market arbitrage checks is an open-source MCP server for finance and payments. It runs locally on your machine and is started by your AI client.

Oracle3 gives agents a research and simulation workflow for prediction markets. Thirteen MCP tools find contracts, read their resolution rules, quotes and order books, calculate venue fees, and run five kinds of relationship checks across related contracts, returning a candidate basket, per-leg fees and a calculated net edge. Orders are simulated in a local paper ledger kept in a JSON file.

The README's synthetic example shows the idea: with an implication between two contracts, a $5.00 gross edge minus $3.4125 in fees leaves $1.5875, conditional on the supplied quotes. Public market reads need no exchange credentials. Real-order execution is separate and lives in the CLI and Python engine, which also supports strategy development with risk controls and paper or live venue adapters.

The project also ships six reusable agent Skills for market research, strategy authoring and operations. It installs from a development checkout or PyPI and needs Python 3.10 or later.

Install

Run in a terminal

claude mcp add oracle3 -- uvx oracle3

Add to claude_desktop_config.json

{
  "mcpServers": {
    "oracle3": {
      "command": "uvx",
      "args": [
        "oracle3"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "oracle3": {
      "command": "uvx",
      "args": [
        "oracle3"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "oracle3": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "oracle3"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "oracle3": {
      "command": "uvx",
      "args": [
        "oracle3"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "oracle3": {
      "command": "uvx",
      "args": [
        "oracle3"
      ]
    }
  }
}

Configuration

  • ORACLE3_MCP_LEDGERPath of the local paper-trading ledger (default ~/.oracle3/mcp_paper_ledger.json)

Tools this server provides (10)

Taken from the project documentation; the running server may differ.

search_markets
Discover markets and contracts.
get_market
Get a contract's metadata and resolution rules.
get_quote
Get prices and fee information.
get_orderbook
Show order book depth.
check_constraint
Check a stated relationship between contracts for net edge.
check_constraint_live
Run a relationship check against live quotes.
trading_fee
Calculate venue trading fees.
fair_value
Compute parameterised model values.
paper_order
Place a simulated order in the local ledger.
paper_portfolio
Show simulated holdings.

What it does

Market discovery
Searches contracts and reads their details and resolution rules.
Quotes and order books
Shows prices, fee information and displayed depth.
Constraint checks
Runs five relationship checks and reports net edge after fees.
Fee calculator
Estimates trading fees and parameterised fair values.
Paper trading
Simulates entries and holdings in a local ledger.
Python engine
Builds strategies with risk controls and venue adapters.

Good for

  • Compare resolution rules across Kalshi and Polymarket contracts
  • Check whether related contracts leave a fee-adjusted edge
  • Practise a strategy with paper orders before any live use
  • Build and test a prediction-market strategy in Python

Pricing

Open source

Open-source project; exchange fees apply separately

Prices change. Check yichengyang-ethan.github.io before you buy.

Oracle3 prediction-market arbitrage checks FAQ

Does MCP place real orders?

No. MCP provides data, calculations and a local paper ledger. Real orders use the separate CLI and Python engine.

Do I need exchange credentials?

Not for public market reads.

How many tools are there?

13 MCP tools and 6 agent Skills.

What does it need?

Python 3.10 or later.

Is Oracle3 prediction-market arbitrage checks free?

Yes. Oracle3 prediction-market arbitrage checks is open source, so you can use and self-host it for free.

How do I add Oracle3 prediction-market arbitrage checks to Claude Code?

Run this command in your terminal: claude mcp add oracle3 -- uvx oracle3

What are the best alternatives to Oracle3 prediction-market arbitrage checks?

The closest alternatives to Oracle3 prediction-market arbitrage checks are Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, Solana MCP by Vybe, Tapetide - Indian Stock Market MCP and World Monitor.

Alternatives to Oracle3 prediction-market arbitrage checks

  1. Open-source analytics app where an AI agent builds charts and dashboards from your data

    • Remote
    • Open source
    • 7.7K GitHub stars
    • Analytics and data
  2. Hosted MCP connector for managing Google Ads, Meta Ads, GA4 and Search Console from chat

    • 8 tools
    • Remote
    • Freemium
    • 8.1K GitHub stars
    • Marketing and SEO
  3. Remote MCP server exposing the Vybe Solana API to AI clients and code editors

    • 44 tools
    • Remote
    • 2.8K GitHub stars
    • Finance and payments
  4. Indian NSE and BSE stock market data for AI assistants through 48 MCP tools

    • 17 tools
    • Remote
    • Freemium
    • 150 GitHub stars
    • Finance and payments
  5. World Monitor

    MCP server

    Live geopolitical intelligence dashboard with an API and MCP server for agents

    • Remote
    • Freemium
    • Analytics and data
  6. Budget Tracker

    MCP server

    Local MCP expense ledger that logs receipts and mileage, splits VAT and prepares rebilling

    • 14 tools
    • Remote
    • Freemium
    • 2 GitHub stars
    • Finance and payments