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)
An equipment maintenance schedule and log: service entries, costs, and what is due or overdue.
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。
MCP_LICENSE_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 …”。
链接指向作者自己的文件;我们不托管也不重新打包它。打开文件后,Claude Desktop 会进行安装。
在终端中运行这条命令
运行前请先填写空着的值。
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。
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8日
注册名称已通过 GitHub 账户 theluckystrike 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:MCP_LICENSE_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(0.22.5)于25天前进入注册中心。
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 MCPB。
同一分类中有:Spec Driven Development, Remote Desktop Commander, Tethered Trust
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.
Fit, compatibility and recall checks for AI agents before they buy: verdicts, evidence, receipts.
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。
In the [official MCP Registry](https://registry.modelcontextprotocol.io/v0.1/servers/io.github.theluckystrike%2Fmaintenance-log/versions/latest) (io.github.theluckystrike/maintenance-log). An mcp maintenance log for the workshop, the rental flats, the van fleet or the studio: one local register of every piece of equipment you look after, and the work that has been done on it. Add each asset once, name, serial or asset tag, where it lives, then log each service or repair as it happens: the date, what was done, what it cost in cents, who did it, and when the next service falls due, either as a date or as an interval in days. Ask what is due and the report answers what is overdue and what comes up in the next N days, computed from the stored dates at the moment you ask, so it can never go stale. Per-asset history keeps the chronological log with its total spend, and export hands the record over as CSV for a spreadsheet or a Markdown summary per asset. Everything stays on this machine; there is no account and no network call.
Built by theluckystrike.
npm publish for @theluckystrike/mcp-maintenance-log is pending, so npx -y @theluckystrike/mcp-maintenance-log returns 404 today. Until then, a clone+build is the working path.
macOS ~/Library/Application Support/Claude/claude_desktop_config.json, Windows %APPDATA%\Claude\claude_desktop_config.json:
{
"mcpServers": { "maintenance-log": { "command": "npx", "args": ["-y", "@theluckystrike/mcp-maintenance-log"] } } }
claude mcp add maintenance-log -- npx -y @theluckystrike/mcp-maintenance-log
~/.cursor/mcp.json (global) or .cursor/mcp.json (project), same entry as Claude Desktop.
| tool | what it does |
|---|---|
asset_add | Add an asset: name, serial or asset tag, location, notes, currency. Returns AST-YYYY-NNNN. A duplicate tag is refused |
maintenance_log | Log work on an asset: the day, what was done, cost in cents, technician, and the next due date -- as next_due or as interval_days, never both |
maintenance_due | The due report: what is overdue and by how many days, what falls due within N days (default 30), what is scheduled later, what has no schedule. Pro |
asset_history | One asset in full: every entry in chronological order, total spend in integer cents, spend per technician, the schedule it currently lives under |
maintenance_export | Entries in a date range as CSV (free) or a Markdown summary per asset (Pro) |
asset_remove | Remove an asset. One carrying a log is refused unless confirm: true, naming what would be lost. The AST number is never reissued |
license_status / license_activate | Free or Pro, and the key |
| Free | Pro | |
|---|---|---|
| Assets on the register | 3 | Unlimited |
| Logging work, cost and technician | Yes | Yes |
| Per-asset history and total spend | Yes | Yes |
| CSV export of a date range | Yes | Yes |
| Due report: overdue and due-within-N-days | -- | Yes |
| Markdown summary per asset | -- | Yes |
The record is never metered. Three assets with full logging and CSV export is a real working register for a one-machine shop, and nothing you have logged is ever held back. What Pro lifts is how many assets are on the register, and adds the due report and the Markdown summaries.
Get Pro: https://mcp.zovo.one/buy/maintenance-log, $19 one-time for this server, or $39 for the bundle.
Every date is stored as ISO 8601 (YYYY-MM-DD). Whether a service is overdue is computed from today's date at call time, never stored, so the register cannot go stale. An interval is turned into a date once, when the work is logged: the work date plus the interval, in whole days, in UTC so the answer does not depend on the server's timezone. Every cost is an integer number of cents in the asset's own currency (1200 is 12.00), stored on the entry; a total spend is the sum of the stored entries, so it can never drift from the lines it is made of.
All data stays local, in ${XDG_DATA_HOME:-~/.local/share}/mcp-servers/maintenance-log/. Two files: assets.json, counter.json. Nothing is sent anywhere, there is no account, no API key and no network call in this server at all. License keys are verified offline.
Built by theluckystrike. https://github.com/theluckystrike
Any MCP client (Claude, Cursor, Windsurf, VS Code) can read this repository's documentation directly via GitMCP - no install: