x402-mcp
@speccy/x402-mcp
MCP 服务器,封装了 Speccy 的付费 x402 端点。为任何兼容 MCP 的 AI 智能体提供两个超强工具——预测市场数据和沙箱化的 Python 执行器——而服务器会以 Base 网络上的 USDC 替你支付每次调用的费用。
智能体无需了解 x402。智能体端无需配置钱包。只需安装服务器,把智能体指向它,搞定。
工具
工具 | 描述 | 每次调用费用 |
| 热门 Polymarket 预测市场(按 volume、liquidity 或 startDate 排序) | $0.01 USDC |
| 在隔离的 Docker 沙箱中运行 Python 3.12 代码(无网络、只读文件系统、30 秒超时、64KB 输出上限) | $0.02 USDC |
Related MCP server: Polymarket MCP Server
安装
npm install -g @speccy/x402-mcp配置
设置操作者钱包密钥(即每次调用支付 x402 的钱包):
export SPECCY_MCP_WALLET_KEY="0x..." # operator wallet private key可选覆盖项:
SPECCY_MCP_API_BASE(默认https://api.speccy.cloud)— Polymarket 端点SPECCY_MCP_EXEC_BASE(默认https://exec.speccy.cloud)— 沙箱执行端点
接入兼容 MCP 的智能体
Claude Desktop(claude_desktop_config.json):
{
"mcpServers": {
"speccy-x402": {
"command": "speccy-x402-mcp",
"env": { "SPECCY_MCP_WALLET_KEY": "0x..." }
}
}
}任意 MCP 客户端(stdio):
SPECCY_MCP_WALLET_KEY=0x... speccy-x402-mcp架构
[Agent] → MCP tool call (stdio) → [this server] → x402 paid fetch → [Speccy endpoint on VPS]
→ [CDP facilitator] → on-chain USDC transfer服务器持有一个操作者钱包。每次调用都会从该钱包发起一笔 x402 支付。结算提醒会通过 Notifier 机器人(与 HTTP 端点使用同一后端)发送到操作者的 Telegram。
备注
钱包密钥绝不会离开操作者的机器(通过环境变量提供,而非配置文件)。
在 Base 主网(
eip155:8453)上为钱包充值 USDC,并充值少量 ETH 作为保证金(仅在钱包进行非 x402 转账时才会用到)。目前仅支持在 Base 主网上结算。测试网 / 其他链需要
X402_NETWORK环境变量以及匹配的 facilitator。每次调用都是独立且无状态的——没有会话,服务器上不保存任何状态。
源码 / 问题
源码:github.com/speccy-ai/x402-mcp(占位符)
问题 / 功能请求:在仓库中提交 issue,或通过 Telegram 联系 @SpeccyNotifierbot
许可证
MIT — 由 Philip (Esla) 为 Speccy 编写。
Available Tools
2 toolsexec_pythonA
Run Python code in an isolated Docker sandbox. Costs $0.02 USDC per execution. No network, read-only filesystem, 30s timeout, 64KB output cap.
Inputs: code (required).
get_prediction_marketsB
Get top Polymarket prediction markets. Costs $0.01 USDC per call (x402 settled in background).
Inputs: sort, limit.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools are completely distinct in purpose: one fetches Polymarket prediction markets, the other executes Python code. There is no realistic risk of an agent confusing them.
Both tool names follow the same lowercase snake_case verb_noun pattern: get_prediction_markets and exec_python. The naming is predictable and consistent.
Two tools is on the thin side, and the tools are unrelated, making the server feel like a loose collection rather than a focused toolkit. The count is not unreasonable for a paid utility server, but it is borderline.
The tools have no shared domain and each is a single isolated operation. get_prediction_markets only returns top markets with no drill-down or follow-up actions, and exec_python is a one-off sandbox execution primitive. The broader x402 workflow is unclear, leaving significant gaps.
Maintenance
Related MCP Connectors
Market data and web intelligence for AI agents, paid per call in USDC on Base via x402.
Market and on-chain crypto data for AI agents. Pay per call in USDC on Base (x402).
17 pay-per-call web, finance, legal and onchain tools for AI agents. USDC on Base via x402/MPP.
Pay-per-call crypto market intelligence for AI agents. USDC on Base via x402.
Related MCP Servers
- AlicenseAqualityCmaintenanceEnables AI agents to discover and analyze prediction markets, execute trades, and manage positions on Polymarket via the Model Context Protocol.786 npm22MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to query Polymarket data such as top wallets, live trades, market details, and smart money flows using natural language through MCP tools.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to access crypto prices, DeFi yields, Polymarket data, Base chain info, and security scans with pay-per-call via USDC on Base mainnet.MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to autonomously trade, analyze, and manage positions on Polymarket prediction markets with 45 tools, real-time WebSocket monitoring, and enterprise-grade safety features.MIT