orcatrace-mcp

orcatrace-mcp

MCP server for OrcaTrace, a pay-per-call Polymarket intelligence feed for AI agents. It provides tools to track smart money, whale entries with calibration, market repricings, and digests across 42K prediction markets, with payments settled via x402 on Base.

Category
访问服务器

README

orcatrace-mcp

MCP server for OrcaTrace — pay-per-call Polymarket intelligence for AI agents. Track smart money and see what others miss across 42K prediction markets: quality-gated repricings with order-book microstructure, tracked-whale entries annotated with each wallet's hold-to-resolution calibration (follow or fade), AI-detected predetermined-outcome markets, on-demand single-market research, and 4-hourly synthesized digests — one unified, index-polled feed.

Pay per call with USDC on Base via x402no account, no API key.

  • 5 free tools — no wallet, no API key.
  • 8 paid tools — flat per-call USD prices ($0.01–$1.00), settled in USDC on Base Mainnet via x402.

The whale track record is free and verifiable. get_track_record publishes the aggregate hold-to-resolution win rate, follow/fade ROI, per-odds-band breakdown, and top wallets with full public addresses — so you can cross-check every claim on Polymarket before you pay. Preview every paid response shape free with get_sample.


Quick start

Free tier (no wallet)

{
  "mcpServers": {
    "orcatrace": {
      "command": "npx",
      "args": ["-y", "orcatrace-mcp"]
    }
  }
}

You get get_index, get_pulse, get_digest_brief, get_sample, get_track_record.

Paid tier (with wallet)

Add an EVM wallet private key (0x-prefixed) that holds USDC on Base Mainnet:

{
  "mcpServers": {
    "orcatrace": {
      "command": "npx",
      "args": ["-y", "orcatrace-mcp"],
      "env": { "PRIVATE_KEY": "0x..." }
    }
  }
}

The paid tools then settle each call automatically.

Install snippets

Claude Code:

claude mcp add orcatrace -- npx -y orcatrace-mcp
# with a wallet:
claude mcp add orcatrace --env PRIVATE_KEY=0x... -- npx -y orcatrace-mcp

Claude Desktop / Cursor: add the JSON block above to your MCP config (claude_desktop_config.json or .cursor/mcp.json).


Tools

Free tools

Tool Endpoint Description
get_index GET /v1/index Current feed index — poll target to detect new items (~10min cadence).
get_pulse GET /v1/pulse Top-3 24h repricings (noise-filtered) + a digest teaser — the free hook.
get_digest_brief GET /v1/digest/brief Digest teaser: title + summary, no analysis (full digest is get_digest).
get_sample GET /v1/sample Static samples of every paid response shape (signal, whales, digest).
get_track_record GET /v1/track-record Verifiable whale-calibration scorecard: win rate, follow/fade ROI, per-odds-band breakdown, top-10 wallets with full public addresses.

Paid tools

Tool Endpoint Price (USD) Description
get_latest GET /v1/latest $0.01 Most recent feed item — mover, whale entry (with calibration), or predetermined signal.
get_signal GET /v1/signal $0.01 One feed item: quality-gated mover (+microstructure), named whale entry, or predetermined signal. Optional index.
get_signals GET /v1/signals $0.03 Batch of up to 20 feed items. Optional since, count, type (mover | whale_entry | predetermined).
get_resolving GET /v1/resolving $0.02 Markets resolving within withinHours (1–72): odds, contested/converged, liquidity, microstructure. Optional minLiquidity, includeNoise.
get_digest_history GET /v1/digest/history $0.05 Historical digests (up to 7 days, 4h cadence). Optional hours, limit.
get_whales GET /v1/whales $0.05 Whale calibration table: per-wallet win rates by odds band, mirror/fade ROI, follow/fade labels (min 30 resolved entries).
get_digest GET /v1/digest $0.10 Full Polymarket Intelligence Digest (every 4h): what repriced and why across 42K markets — synthesized analysis.
get_research GET /v1/research $1.00 Single-market deep-dive: 7-day price action + order book + whale positioning + news, synthesized into a thesis. Requires market slug.

Prices are flat per call — no bundles, no entitlements. Confirm live prices any time via the free get_sample / the site.


How payments work (x402)

OrcaTrace speaks the x402 exact scheme. The API's 402 advertises a Base (USDC, EIP-3009) rail; this MCP settles on Base:

  1. The MCP requests a paid endpoint. With no payment, the API replies HTTP 402 with the accepted terms.
  2. If PRIVATE_KEY is set, the MCP settles the call via the standard @x402/fetch V2 client and retries. If no key is set, the MCP returns the 402 price terms so you can see the cost without paying.
  3. The API returns the data plus a payment block.

You are never charged for an error. The feed and digest routes are compute-first / settle-after — the full answer is computed before you are charged, and the digest returns 503 (no charge) rather than serving stale. The proxied-compute routes (get_resolving, get_whales, get_research) settle-early with a guaranteed refund — if the API charges but cannot deliver, the payment is refunded.

No accounts, no API keys — just a wallet with USDC on Base Mainnet.


Environment variables

Variable Required Description
PRIVATE_KEY No 0x-prefixed wallet key with USDC on Base Mainnet. Required to pay the 8 paid tools on a live API. Free tools work without it; without a key the paid tools surface the 402 price terms.
ORCATRACE_API_URL No Override the API base URL (default https://api.orcatrace.dev).

Disclaimer

OrcaTrace provides informational prediction-market data and analysis aggregated from public sources. Whale calibration measures entries held to resolution, not trader P&L. Nothing here is financial advice, and everything is provided as-is without warranty. See https://api.orcatrace.dev/terms.txt.

Docs for agents: https://api.orcatrace.dev/llms.txt · full: https://api.orcatrace.dev/llms-full.txt


License

MIT © jcislo

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选