polymarket-mcp
An MCP server and Python toolkit that provides AI agents with real-time tools for Polymarket prediction markets, including liquidity scanning, arbitrage detection, and slippage estimation. It also offers advanced wallet intelligence, portfolio risk calculation, and probabilistic reasoning to enhance market analysis and strategy.
README
polymarket-mcp
<!-- mcp-name: io.github.whitmorelabs/polymarket-mcp -->
Prediction market intelligence for AI agents. Wallet scores, arbitrage detection, liquidity analysis, and real-time price feeds for Polymarket — backed by proprietary data from 195+ curated on-chain wallets.
What can it do?
Ask your AI agent:
- "Find all Polymarket markets where YES + NO prices don't add up to 100%"
- "What's the slippage if I buy $500 of YES on the Bitcoin $100k market?"
- "Show me the top 20 wallets on Polymarket by score and PnL"
- "Which wallets co-trade alongside this whale? Find the shadow network."
- "What's the liquidity grade for this market — is it safe to enter?"
- "Calculate my portfolio risk across these 5 positions"
- "What's the probability that Bitcoin hits $150k by June?"
- "Scan today's crypto news and match to open Polymarket markets"
Install
pip install polymarket-trading-mcp
Or run directly:
uvx --python 3.11 polymarket-trading-mcp
MCP Config
<details> <summary><strong>Claude Desktop</strong></summary>
Add to your claude_desktop_config.json:
{
"mcpServers": {
"polymarket": {
"command": "polymarket-mcp",
"args": ["--stdio"]
}
}
}
</details>
<details> <summary><strong>Cursor / Windsurf / other MCP clients</strong></summary>
Run as SSE server:
polymarket-mcp
# Starts on http://127.0.0.1:8000/sse
Run as streamable HTTP:
polymarket-mcp --http
# Starts on http://127.0.0.1:8000/mcp
Then point your client at http://127.0.0.1:8000/sse or http://127.0.0.1:8000/mcp.
</details>
How it compares
| Feature | polymarket-mcp | Manual CLOB calls | py-clob-client |
|---|---|---|---|
| MCP integration | Yes | No | No |
| Slippage estimation | Yes | DIY | Partial |
| Arbitrage scanning | Yes | DIY | No |
| Wallet intelligence | Yes | No | No |
| Price feed (multi-source) | Yes | No | No |
| Probability engine | Yes | No | No |
| Portfolio risk math | Yes | No | No |
| AI agent ready | Yes | No | No |
Tools
Free Tools
<details> <summary><strong>estimate_slippage</strong> — Estimate fill cost at different order sizes</summary>
Walk the CLOB order book and calculate average fill price for a given position size.
estimate_slippage(
market_id="71321045679252212594626385532587807886473352341859285086426388424842616389328",
side="BUY",
position_size_usd=500.0
)
Returns: best_price, avg_fill_price, slippage_pct, liquidity_available, recommendation (go/caution/no-go).
</details>
<details> <summary><strong>scan_market_liquidity</strong> — Liquidity depth, spread, and grade</summary>
Scan a market's liquidity depth with spread and depth at 1%/2%/5% from mid.
scan_market_liquidity(market_id="71321045679252212594626385532587807886473352341859285086426388424842616389328")
Returns: best_bid, best_ask, spread_pct, depth buckets, liquidity_grade (A/B/C).
</details>
<details> <summary><strong>find_arbitrage_gaps</strong> — Scan all markets for YES+NO mispricing</summary>
Scan all active Polymarket markets for YES + NO price sums that don't equal 1.00.
find_arbitrage_gaps(min_gap_pct=2.0)
Returns: markets with gaps, gap size, and direction to exploit (BUY_BOTH or SELL_BOTH).
</details>
<details> <summary><strong>get_crypto_price</strong> — Real-time prices from Binance and CoinGecko</summary>
Real-time crypto price from Binance and/or CoinGecko in parallel, with latency metadata.
get_crypto_price(symbol="BTC", sources="binance,coingecko")
Returns: per-source prices, average, max deviation between sources.
</details>
<details> <summary><strong>get_contract_status</strong> — Time remaining and resolution status</summary>
Time remaining and resolution status for active Polymarket contracts. BTC 5-min/15-min contracts use slug timestamp parsing for exact expiry.
get_contract_status(market_id="btc-updown-5m-1711234567")
get_contract_status(category="crypto") # scan all short-duration contracts
Returns: time_remaining_seconds, time_remaining_human, current_odds.
</details>
<details> <summary><strong>estimate_probability</strong> — Structured probability reasoning for any question</summary>
Estimate the probability of any event using structured reasoning. Base rates, key factors, timeframe analysis — all in one structured JSON response.
estimate_probability(question="Will Bitcoin reach $150k before end of 2025?")
Returns: estimated_probability (0-1), confidence_level, factors_for, factors_against, base_rate_note, reasoning.
</details>
<details> <summary><strong>calculate_portfolio_risk</strong> — Pure-math risk metrics across positions</summary>
Pure-math risk metrics for a portfolio of prediction market positions. No external API calls.
calculate_portfolio_risk(positions='[{"market_id": "abc123", "side": "YES", "size": 200, "entry_price": 0.65}]')
Returns: total_exposure, max_loss, max_gain, diversification_score, kelly_warnings, rebalancing_suggestions.
</details>
<details> <summary><strong>get_scoring_performance</strong> — Metadata and dataset stats (free)</summary>
Returns scoring methodology, current dataset stats, and pricing. Call this first to evaluate whether the paid data is worth it. No API key required.
</details>
Paid Tools (API key required)
7-day free trial on signup. After trial: pay-per-request in USDC on Polygon. Volume discount: 50% off above 1000 requests/day.
| Tool | Cost |
|---|---|
get_shadow_wallets |
$0.05/request |
get_top_wallets |
$0.05/request |
get_wallet_score |
$0.05/request |
get_wallet_pnl |
$0.05/request |
get_wallet_clusters |
$0.05/request |
get_btc_short_markets |
$0.01/request |
analyze_market_sentiment |
$0.05/request |
<details> <summary><strong>get_shadow_wallets</strong> — Find wallets that co-trade with a whale</summary>
Find wallets that co-trade with a seed wallet across the same Polymarket categories.
get_shadow_wallets(seed_wallet="0xabc...", api_key="your-key")
Returns: up to 20 shadow wallets ranked by co-trading intensity, shared categories, win rates.
</details>
<details> <summary><strong>get_top_wallets</strong> — Top-ranked wallets by score and PnL</summary>
Top-ranked Polymarket wallets from MoonMirror's scoring engine (PnL + consistency + specialization - drawdown penalty).
get_top_wallets(api_key="your-key", limit=50, min_score=0.5)
</details>
<details> <summary><strong>get_wallet_score</strong> — Full scoring breakdown for any wallet</summary>
Full scoring breakdown for a specific wallet address.
get_wallet_score(address="0xabc...", api_key="your-key")
Returns: score components, win rate, trade count, data freshness.
</details>
<details> <summary><strong>get_wallet_pnl</strong> — Realized PnL with per-category breakdown</summary>
Realized PnL for a wallet with per-category breakdown. Pre-computed from Goldsky subgraph.
get_wallet_pnl(address="0xabc...", api_key="your-key")
</details>
<details> <summary><strong>get_wallet_clusters</strong> — Wallets independently co-trading with multiple top wallets</summary>
Wallets that independently appear alongside 3+ different top-scoring wallets — stronger signal than a single shadow match.
get_wallet_clusters(api_key="your-key", min_co_trades=3)
</details>
<details> <summary><strong>get_btc_short_markets</strong> — Active BTC 5-min and 15-min contracts</summary>
Active BTC 5-min and 15-min Polymarket contracts with token IDs and current midpoint prices. Cached, refreshed every 60 seconds.
get_btc_short_markets(api_key="your-key")
</details>
<details> <summary><strong>analyze_market_sentiment</strong> — Structured probabilistic analysis</summary>
Structured probabilistic analysis for a prediction market question using Claude's reasoning.
analyze_market_sentiment(question="Will Ethereum ETF hit $10B AUM by Q3 2025?", api_key="your-key")
</details>
Pricing
| Tier | Cost | Access |
|---|---|---|
| Free tools | $0 | Slippage, liquidity, timer, price feeds, arbitrage, risk calculator |
| Paid tools | $0.05/request | Wallet intelligence, sentiment analysis |
| Bundle | pip install polymarket-trading-mcp | All 10 tools |
Standalone packages also available: polymarket-slippage, polymarket-liquidity, polymarket-arb-detector, polymarket-price-feed, polymarket-sentiment, polymarket-timer, polymarket-news, polymarket-onchain
Why this server?
Every other Polymarket MCP server queries public APIs. We have proprietary wallet intelligence — curated scores, real PnL data from Goldsky, and shadow wallet networks discovered through counterparty mining. Data you can't get anywhere else.
The paid wallet tools are backed by MoonMirror, a Polymarket wallet intelligence engine. MoonMirror continuously monitors wallets on-chain, scores them on PnL, consistency, and market specialization, and pre-computes shadow wallet relationships.
All wallet queries return in < 100ms because everything is pre-computed. No blocking subgraph calls.
Configuration
All settings via environment variables:
# Server
HOST=127.0.0.1
PORT=8000
# DB paths (for self-hosted wallet intelligence)
MOONMIRROR_DB_PATH=/path/to/moonmirror.db
SERVICES_DB_PATH=/path/to/services.db
# Pricing
PRICE_WALLET_INTEL=0.05
PRICE_MARKET_DATA=0.01
TRIAL_DAYS=7
# Rate limiting
RATE_PER_KEY=5
RATE_GLOBAL=20
<details> <summary><strong>Admin CLI</strong></summary>
# Create an API key
polymarket-mcp --create-key "customer-name"
# Top up a key's USDC balance
polymarket-mcp --top-up YOUR_KEY 10.0
</details>
github.com/whitmorelabs/polymarket-mcp — issues, PRs, and stars welcome.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.