SimpleFunctions

Compare · MCP servers

Prediction Market MCP Servers,
9 Compared.

Nine Model Context Protocol servers that expose Kalshi, Polymarket, or related prediction-market tools to AI agents. Each compared by tool count, venue coverage, install command, license, auth model, and best-fit use case — verified from public sources only.

Servers compared

9

Multi-venue

4

No-auth read

7

SimpleFunctions tools

101

Quick recommendations

If you want X, install Y.

  • Cross-venue normalized data + indicators + calibration in production → SimpleFunctions MCP
  • Cross-venue read-only for personal use, no auth → JamesANZ / prediction-market-mcp
  • Polymarket-only thicker tool surface → guangxiangdebizi / PolyMarket-MCP
  • Solana-native venue (Baozi) → Baozi MCP Server
  • Polymarket whale-wallet attribution → PolyVision MCP
  • Polymarket-only minimal install → TanmayDhobale / polymcp
  • Reference implementation to fork → berlinbra / polymarket-mcp
  • Hosted autonomous-trading agent (not self-hostable MCP) → Clawmarket

Side by side

Comparison table.

ServerToolsVenuesLicenseAuth
SimpleFunctions MCP101Kalshi, Polymarketproprietarymixed
JamesANZ / prediction-market-mcp~10Polymarket, PredictIt, KalshiMITnone
Baozi MCP Server68Baozi (Solana)proprietarywallet
guangxiangdebizi / PolyMarket-MCP~14PolymarketMITnone
TanmayDhobale / polymcp~6PolymarketMITnone
prediction-markets-mcp (npm)~8Polymarket, KalshiMITnone
PolyVision MCP~5Polymarketunknownnone
berlinbra / polymarket-mcp~4PolymarketMITnone
Clawmarket (Polymarket + Kalshi autonomous agent)agent (not a public MCP)Polymarket, Kalshiproprietarywallet

Tool counts marked with `~` are approximate from public READMEs at time of review (2026-05) and may have changed. SimpleFunctions's current MCP count is published at /api/internal/statistics and verified against src/app/api/mcp/[transport]/route.ts in the spfunctions repo.

Detail · 9 servers

Each server, in one paragraph.

SimpleFunctions MCP

https://simplefunctions.dev/api/mcp/mcp

KalshiPolymarket

Cross-venue Kalshi + Polymarket reads, screener with computed indicators (IY/CRI/LAS/EE), causal-tree thesis system, calibration scores, world-state snapshot, agent guide, and authenticated intent execution. The only commercially-supported multi-venue MCP server with a published SLA-style uptime page.

Best for: Production AI agents that need normalized cross-venue prices + indicators + calibration in one MCP install, plus thesis decomposition and autopilot intents.

claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp
JamesANZ / prediction-market-mcp

https://github.com/JamesANZ/prediction-market-mcp

PolymarketPredictItKalshi

Open-source TypeScript MCP server aggregating Polymarket, PredictIt, and Kalshi public endpoints. Listed on PulseMCP. No API keys required for the read tools. Hobby-grade — no SLA, no auth, no derived signals.

Best for: Local prototyping or personal Claude Desktop / Cursor sessions where you need quick read-access across three venues without registration.

npx -y prediction-market-mcp
Baozi MCP Server

https://www.npmjs.com/package/@baozi.bet/mcp-server

Baozi (Solana)

Single-venue MCP for the Baozi Solana prediction market. Largest tool count on this list (68) but covers only one venue and ties to the Baozi product roadmap.

Best for: Builders specifically targeting Solana-native prediction markets through the Baozi ecosystem and willing to integrate wallet auth.

npx -y @baozi.bet/mcp-server
guangxiangdebizi / PolyMarket-MCP

https://github.com/guangxiangdebizi/PolyMarket-MCP

Polymarket

Comprehensive Polymarket-only MCP exposing markets, orderbook, trades, holders, and news endpoints in Python. Active GitHub repo with Chinese + English README.

Best for: Polymarket-only research workflows that need a thicker tool surface than the simple aggregator MCPs.

git clone + python -m polymarket_mcp
TanmayDhobale / polymcp

https://github.com/TanmayDhobale/polymcp

Polymarket

Minimal Polymarket-only MCP server in Python. Smaller scope than the guangxiangdebizi repo — read-only market endpoints.

Best for: Reading the Polymarket public API from Claude Desktop with the smallest possible install footprint.

git clone + uv run polymcp
prediction-markets-mcp (npm)

https://www.npmjs.com/package/prediction-markets-mcp

PolymarketKalshi

Generic npm-published MCP server bundling Polymarket + Kalshi public reads. Overlaps in scope with the JamesANZ project; check published version + commit recency before installing.

Best for: Quick install via npx without cloning a repo — useful for ephemeral Claude Code sessions.

npx -y prediction-markets-mcp
PolyVision MCP

https://github.com/polyvision/polyvision-mcp

Polymarket

Wallet-analysis-focused Polymarket MCP — surfaces top traders, holdings, and PnL on specific Polymarket addresses rather than market data per se.

Best for: Watching specific Polymarket whale wallets or studying trader-attribution patterns from an agent.

git clone + python -m polyvision_mcp
berlinbra / polymarket-mcp

https://github.com/berlinbra/polymarket-mcp

Polymarket

Early Polymarket MCP project; smaller surface than guangxiangdebizi but a clean reference implementation for first-time MCP authors.

Best for: Reading the source as a starter template if you want to build your own Polymarket MCP server.

git clone + python -m polymarket_mcp
Clawmarket (Polymarket + Kalshi autonomous agent)

https://www.useclawmarket.com

PolymarketKalshi

Third-party autonomous-trading agent product over Polymarket + Kalshi. Distributed as a hosted agent rather than a self-hostable MCP server.

Best for: Users who want a turnkey trading agent rather than a programmable MCP surface to plug into their own Claude Code / Cursor session.

Note: Different product from SimpleFunctions' /openclaw landing page (which references the Claude Skill packaging for SimpleFunctions tools, not the Clawmarket service).

see useclawmarket.com onboarding

FAQ

Common questions.

What is an MCP server for prediction markets?

A Model Context Protocol (MCP) server exposes a set of typed tools to AI agents — Claude Code, Cursor, Claude Desktop, and other MCP-compatible clients. For prediction markets that means tools like get_markets, get_orderbook, get_calibration, screen_indicators, create_thesis. The agent calls them by name; the MCP server hits the underlying venue (Kalshi, Polymarket, etc.) and returns normalized JSON. Without MCP, the agent would have to be taught each REST endpoint manually.

How do I install an MCP server in Claude Code?

For SimpleFunctions: run `claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp`. For npm-published servers: `npx -y <package-name>` and add the resulting command to your `claude_desktop_config.json` mcpServers map. For git-based servers: clone the repo, follow its README for the launch command, and reference that command in the same config. Restart your Claude client after editing the config.

Why does SimpleFunctions ship more tools than most single-venue MCP servers?

Most third-party MCPs cover one venue's read endpoints. SimpleFunctions covers Kalshi + Polymarket reads plus seven additional surfaces an agent typically wants on top of raw markets: a screener over computed indicators (implied yield, cliff risk, liquidity availability score), causal-tree thesis decomposition + signal injection, live calibration scores, world-state snapshots and deltas, news-stream subscriptions, a public agent-guide tool, and authenticated intent execution. The current count is published at /api/internal/statistics so the page does not rely on stale marketing copy.

Are these MCP servers free?

Most of the third-party servers on this list are MIT-licensed, no-auth, and free for personal use — the cost is just whatever rate limits the underlying venue (Polymarket public API, Kalshi public REST) imposes on unauthenticated traffic. SimpleFunctions reads through the public MCP endpoint are also free with no auth required; authenticated tools (thesis system, intents, autopilot) are free up to 15M LLM tokens, then pay-per-token. Baozi MCP and Clawmarket are closer to product subscriptions.

Which MCP server should I install if I want cross-venue arbitrage?

SimpleFunctions is currently the only MCP server that exposes a `cross_venue_pairs` tool returning matched contracts across Kalshi and Polymarket with live spreads. Other multi-venue servers (JamesANZ, prediction-markets-mcp) return separate venue feeds — you would need to write the matching logic yourself in the agent loop. For arbitrage workflows, SimpleFunctions's built-in pair table saves the canonical-question-normalization step.

Is Clawmarket the same as SimpleFunctions /openclaw?

No. Clawmarket (useclawmarket.com) is a third-party autonomous-trading agent product over Polymarket and Kalshi. SimpleFunctions /openclaw is a landing page for the Claude Skill packaging of SimpleFunctions's MCP tools — it is the SimpleFunctions product, not the Clawmarket service. The two are unrelated commercial offerings; we list Clawmarket here because it is sometimes confused with SimpleFunctions's open-source MCP surface.

How do I evaluate an MCP server before installing it?

Three checks: (1) read the source — is it doing what the README claims, or piping a third-party API through unaudited code? (2) check commit recency — MCPs depending on Polymarket / Kalshi public endpoints can break silently when those APIs change shape; an unmaintained server is a debugging cliff. (3) check the auth model — anything requiring wallet keys deserves a code audit before the first run.

Can I run multiple MCP servers at the same time?

Yes. Claude Code, Claude Desktop, and Cursor all support multiple MCP servers in parallel; each registers its tools under a namespaced prefix. A common pattern is SimpleFunctions for cross-venue normalized data plus a venue-specific MCP (e.g. guangxiangdebizi/PolyMarket-MCP) for tools that only that venue's native API exposes. Tool name collisions are resolved by the prefix.

Install SimpleFunctions MCP.

101 tools across Kalshi + Polymarket, normalized prices, indicator screener, calibration, thesis system, world-state snapshot. One line.

claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp