Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
作者提供的说明(英文原文)
Pre-trade DeFi intelligence: 24 paid x402 tools incl. 6 synthesis verdicts, USDC on Base.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
HYPERD_WALLET_PRIVATE_KEY标为机密的值是你账户的钥匙。千万不要分享给别人,也不要粘贴到聊天里。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"hyperd": {
"command": "npx",
"args": [
"-y",
"hyperd-mcp"
],
"env": {
"HYPERD_WALLET_PRIVATE_KEY": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add hyperd --env HYPERD_WALLET_PRIVATE_KEY=your-value -- npx -y hyperd-mcp运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"hyperd": {
"command": "npx",
"args": [
"-y",
"hyperd-mcp"
],
"env": {
"HYPERD_WALLET_PRIVATE_KEY": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"hyperd": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"hyperd-mcp"
],
"env": {
"HYPERD_WALLET_PRIVATE_KEY": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月6日
注册名称已通过 GitHub 账户 hyperd-ai 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
你需要填写:HYPERD_WALLET_PRIVATE_KEY
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(1.1.1)于5个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Spec Driven Development, Remote Desktop Commander, BankBridge
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Works in any MCP agent, including local models. Read-only bank accounts, spending, investments.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Exposes 24 paid hyperD x402 tools — 17 data/risk endpoints, 6 premium synthesis verdicts ($0.30–$1.50), and a multi-call bundle — plus 3 free meta tools (and 3 watch tools when enabled) from api.hyperd.ai. Drop into Claude Desktop, Cursor, Cline, Zed, or any MCP-compatible client.
Each tool call:
api.hyperd.ai endpointYou pay $0.005–$1.50 in USDC on Base per call — data/risk endpoints $0.005–$0.10, premium synthesis verdicts $0.30–$1.50. No subscription. No accounts.
First 5 calls per IP per 24h are free — no wallet, no signup, no API key. Just curl:
curl "https://api.hyperd.ai/api/balance?address=0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045"
Lifetime cap: 10 calls per IP. After that (or when daily quota is exhausted), the endpoint returns HTTP 402 — sign a small EIP-3009 USDC payment on Base via the Python SDK or TypeScript MCP server. The premium synthesis tier ($0.30+) always requires payment — it is not free-tier eligible.
/api/wallet/pnl has a tighter free-tier cap of 1 call/IP/day (heavy upstream).
For testing, use a fresh wallet, not your main one. A simple way:
node -e "const {generatePrivateKey, privateKeyToAccount} = require('viem/accounts'); const pk = generatePrivateKey(); console.log('Address:', privateKeyToAccount(pk).address); console.log('Private key:', pk);"
Send ~$1 of USDC on Base to the printed address.
Edit your Claude Desktop config (location: macOS ~/Library/Application Support/Claude/claude_desktop_config.json, Windows %APPDATA%\Claude\claude_desktop_config.json):
{
"mcpServers": { "hyperd": { "command": "npx", "args": ["-y", "hyperd-mcp"], "env": { "HYPERD_WALLET_PRIVATE_KEY": "0xYOUR_PRIVATE_KEY_HERE" } } } }
Restart Claude Desktop. The hyperD tools appear in the conversation tool list.
Ask Claude in any conversation:
"What's the security score for USDC on Base? Use the hyperd.token.security tool."
Claude calls our endpoint, pays $0.05 from your wallet, returns the result. The whole flow takes ~2 seconds.
Each MCP-compatible client has a similar config file. Same pattern: name the server hyperd, command is npx -y hyperd-mcp, set HYPERD_WALLET_PRIVATE_KEY env var.
For the full MCP client list and per-client config docs see https://modelcontextprotocol.io/clients.
If you want to run from source instead of npx:
git clone https://github.com/hyperd-ai/hyperd-mcp.git
cd hyperd-mcp npm install npm run build
Then in your Claude Desktop config, point at the local build:
{
"mcpServers": { "hyperd": { "command": "node", "args": ["/full/path/to/hyperd-mcp/dist/server.js"], "env": { "HYPERD_WALLET_PRIVATE_KEY": "0xYOUR_PRIVATE_KEY_HERE" } } } }
| Tool | Price | What it does |
|---|---|---|
hyperd.balance.get | $0.01 | Multi-chain ERC-20 + native balance lookup |
hyperd.yield.recommend | $0.05 | Opinionated DeFi yield recommendation |
hyperd.token.info | $0.01 | Aggregated token metadata (CoinGecko + DefiLlama) |
hyperd.token.security | $0.05 | GoPlus-backed security risk score |
hyperd.wallet.risk | $0.10 | Chainalysis sanctions + GoPlus heuristics |
hyperd.protocol.tvl | $0.01 | DefiLlama protocol health |
hyperd.gas.estimate | $0.005 | Gas oracle with tip percentiles |
hyperd.dex.quote | $0.02 | Multi-aggregator best DEX route |
hyperd.wallet.persona | $0.10 | Behavioral wallet classification |
hyperd.contract.audit | $0.10 | Pre-trade contract security composite |
hyperd.governance.summarize | $0.10 | LLM-summarized DAO proposals |
hyperd.sentiment.token | $0.05 | Token sentiment from Farcaster |
hyperd.liquidation.risk | $0.10 | Cross-protocol liquidation risk (Aave V3 / Compound v3 / Spark / Morpho) |
hyperd.wallet.anomaly | $0.10 | Wallet anomaly detection vs 180-day baseline |
hyperd.wallet.pnl | $0.05 | Realized + unrealized P&L (FIFO/LIFO/HCFO) |
hyperd.budget.guardian | $0.01 | Agent USDC spend visibility + cap check |
hyperd.bundle | $0.20 fixed | Multi-call: 1-10 paid GETs in one settlement |
hyperd.watch.create | $3.00 prepay | Subscribe to a continuous liquidation watch (HMAC webhooks) |
hyperd.watch.list | (free for owner) | List your active watches |
hyperd.watch.cancel | (free for owner) | Cancel one of your watches |
Each synthesis tool fans out multiple sub-calls internally and returns a single verdict + inputs + methodology.version + coverage envelope, authored by Claude Haiku 4.5.
| Tool | Price | What it does |
|---|---|---|
hyperd.risk.full_audit | $0.35 | Composed wallet risk: balance + sanctions + persona + contract + mixer. Bands: safe / moderate / elevated / critical. |
hyperd.token.archetype | $0.30 | Token classification: stablecoin / blue_chip / mid_cap / memecoin / wrapped / governance / unverified. |
hyperd.wallet.thesis | $0.50 | Plain-language behavioral thesis: trader / hodler / yield_farmer / mev_bot / compromised / inactive. |
hyperd.wallet.threat_brief | $1.50 | 2-3 sentence security verdict: clean / watch / elevated / critical. |
hyperd.gov.translate | $1.00 | Snapshot/Tally proposal in plain language (1y cache). |
hyperd.yield.allocation | $1.00 | Portfolio split across risk tiers for a given amount (6h cache). |
Full HTTP API docs: https://api.hyperd.ai/api/discover
If you have a 12 or 24-word BIP-39 mnemonic instead of a raw key:
"env": {
"HYPERD_WALLET_MNEMONIC": "word1 word2 ... word12" }
Derives at the standard Ethereum path m/44'/60'/0'/0/0.
HYPERD_API_BASE — default https://api.hyperd.ai. Override for testnet or self-hosted.HYPERD_BASE_RPC — default public Base RPC. Override for higher rate limits.If you don't want to run the stdio server locally, hyperD's full tool catalog is also available as a remote MCP server:
POST https://api.hyperd.ai/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}
Same 24 paid tools, same free-tier quota (5 calls/IP/day, 10 lifetime), same X-Payment auth model for paid calls after quota. No npm install, no per-IDE config — just a URL.
Useful when:
curl before committing to a local installServer card: https://api.hyperd.ai/.well-known/mcp.json.
MIT — same as the parent repo.