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)
Dated formula history, confirmed changes, and evidence for U.S. pre-workout supplements.
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。
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": {
"supprecord": {
"command": "npx",
"args": [
"-y",
"@reneturcios/supprecord",
"mcp"
]
}
}
}在终端中运行这条命令
claude mcp add supprecord -- npx -y @reneturcios/supprecord mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"supprecord": {
"command": "npx",
"args": [
"-y",
"@reneturcios/supprecord",
"mcp"
]
}
}
}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": {
"supprecord": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/supprecord",
"mcp"
]
}
}
}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 CloudCorpRecords 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。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
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(0.1.0)于前天进入注册中心。
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, Docker。
同一分类中有:Spec Driven Development, Remote Desktop Commander, FOGA node
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.
A Finch OG Agent cell as an MCP server: read the world, dry-run and take ERC-8414 tenders by policy.
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。
supprecord is a small local bridge to FormulaSignal's hosted MCP server, which maintains a dated evidence record for a defined, counted set of U.S. pre-workout supplements: what each covered product declares now, whether its formula has changed, and the evidence behind each confirmed change.
The bridge keeps no record of its own. It serves the upstream server's tool list locally, exactly as captured from the live endpoint, and forwards each tool call upstream, returning the result untouched. Reads dominate the surface; the only writes are additions to and removals from the caller's own watchlist.
record.search — answer one bounded natural-language question from the record. Required: query (at most 300 characters).products.resolve — resolve a brand, product name or alias to one canonical covered product. Required: identifier.products.get — the current record for one product: identity, serving configuration, declared ingredients, captured price context and freshness. Required: product_id.products.history — the historical timeline for one product, with each state classified by what it can support. Required: product_id.signals.explain — explain one approved signal: the before state, the after state, the observation window and the evidence. Required: signal_id.signals.list — approved signals in release order, newest first, for incremental polling. All filters are optional: since, until, product_id, brand, dimension, category, limit and cursor.products.compare — compare two or three covered products on consistent criteria. Required: product_ids (2 to 3 ids).products.serving_economics — captured commercial facts and deterministic price arithmetic for one product. Required: product_id.ingredients.research — the dose range used in the selected evidence set for named ingredients, with citations and limitations. Optional: ingredient_ids (up to 5) or product_id, not both.ingredients.regulatory — regulatory records and documented cautions for named ingredients. Optional: ingredient_ids (up to 5) or product_id, not both.signals.snapshot — a bounded summary of approved signals and coverage state for a date window. Optional: from, to, limit and cursor.ledger.list — list the published Category Ledger editions. Takes no arguments.ledger.get — return one Category Ledger edition. Required: edition_id.watchlist.list — the watchlist of the account the key is bound to, with each product's monitoring state. Takes no arguments.watchlist.add — add one covered product to the bound account's watchlist. Required: product_id.watchlist.remove — remove one product from the bound account's watchlist. Required: product_id.watchlist.receipt — the monitoring receipt for one period. Optional: period (YYYY-MM).record.version — the public record version, methodology version, supported category, coverage counts and source freshness. Takes no arguments.FORMULASIGNAL_MCP_URL (optional) — override the upstream endpoint. Defaults to https://formulasignal.com/mcp. (The override hook is this rebuild's own labelled reconstruction.)FORMULASIGNAL_API_KEY (optional) — when set, every upstream request carries an Authorization: Bearer header. When it is not set, no auth header is sent. (This forwarding is also a labelled reconstruction; the original surface was captured keyless.)On the first tool call the client opens a streamable HTTP session with the upstream: a JSON-RPC initialize request (protocolVersion 2025-03-26, clientInfo name supprecord), then a notifications/initialized notification, then the tools/call request carrying the caller's arguments unchanged. When the upstream answers initialize with an Mcp-Session-Id header, that value rides on every later request of the session, which is reused for the life of the process. Replies may arrive as plain JSON or as server-sent events; both are read.
Successful upstream results pass through verbatim, including any isError flag the upstream sets. Failures — a JSON-RPC error, a non-2xx status, an unreadable response, or a request that cannot be completed — come back as an isError result whose text starts with supprecord error: (this rendering is this rebuild's own labelled reconstruction).
Each tool is also a subcommand taking one JSON argument string (omit it for tools that take none). Tool names contain dots and are used exactly as shown:
npx @reneturcios/supprecord record.search '{"query":"How much caffeine is in C4 Original?"}'
npx @reneturcios/supprecord products.resolve '{"identifier":"C4 Original"}' npx @reneturcios/supprecord products.get '{"product_id":"example-id"}' npx @reneturcios/supprecord signals.list '{"limit":10}' npx @reneturcios/supprecord ledger.list npx @reneturcios/supprecord --version
Exit codes: 0 on success, 1 on an isError result, 2 on a usage error.
npx -y @reneturcios/supprecord mcp
MCP client config:
{
"mcpServers": { "supprecord": { "command": "npx", "args": ["-y", "@reneturcios/supprecord", "mcp"] } } }
docker run -i ghcr.io/cloudcorprecords/supprecord
docker run --rm ghcr.io/cloudcorprecords/supprecord --version
The container's default command is mcp, so the first form is a stdio MCP server; any other arguments reach the CLI normally.
npm install
npm run build npm test npm run dev
The test suite runs entirely against a stubbed JSON-RPC upstream — no live network calls.
MIT — see LICENSE.