MyMCPTools

io.github.shibley由 GitHub 账号 shibley 发布v0.1.1更新于 Oct 24, 2026

Find MCP servers and check whether they actually respond, via live handshake probes.

已验证Streamable HTTP可网页运行网页搜索与抓取开发者工具安全与监控

概览

AI 生成的概览

一个托管、只读的 MCP 服务器，用于检索 MCP 服务器目录，并通过实时握手探测确认所列服务器是否真的可响应。

可以做什么
  • 按关键词、分类、客户端集成、安装类型或官方状态检索 MCP 服务器目录，并列出分类与集成及其数量。
  • 查看单个服务器的完整条目：描述、作者、安装命令、仓库、支持的客户端和当前探测结论。
  • 查询单个服务器或整个目录的实时状态：结论、工具数量、握手延迟、协议版本和最近健康时间。
  • 查看探测历史与每日在线率走势，并列出重建的故障事件。
  • 追踪探测之间的工具 schema 与协议版本漂移，并获取整体目录健康统计。
适用场景
适合在挑选要添加的 MCP 服务器时，确认远程服务器确实可达；也适合监控已依赖的服务器是否出现故障或破坏性工具变更。按分类或客户端浏览目录时同样有用。
运行要求
需要访问 mymcptools.com 的远程 Streamable HTTP 端点，无需 API 密钥或认证。另有本地 stdio 适配器，但尚未发布到 npm，需用 Node.js 18+ 从源码构建；可用 MYMCPTOOLS_MCP_URL 覆盖上游端点。
安装前请注意
只读且无状态，不索取任何凭据。探测结论只覆盖远程端点：目录中大多数条目为本地服务器，会显示为 UNPROBEABLE，意思是无法远程检查，而非已损坏。在线率与漂移数据来自运营方自身的探测，并非独立验证。

安装

在 SourceWeft 中

  1. 打开 控制台中的 MyMCPTools，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体，或 SourceWeft 对话中。智能体会读取该服务器的安装指南，先向你展示发布者、端点和登录方式，经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/io.github.shibley%2Fmymcptools/install.md，并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前，先告诉我它的发布者、端点和登录方式，等我确认。不要把 API 密钥写进命令或我会提交的文件；问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "mymcptools": {
      "type": "http",
      "url": "https://mymcptools.com/api/mcp"
    }
  }
}

README

MyMCPTools MCP Server

An MCP server for finding other MCP servers — and, more usefully, for checking whether the one you are about to recommend is actually reachable.

Every uptime figure it returns comes from a real MCP handshake against the server's remote endpoint, not from self-reported metadata in a README.

Backed by mymcptools.com. No API key required.

Install

Remote endpoint (recommended — nothing to install)

The server is hosted. If your client speaks Streamable HTTP, point it straight at:

https://mymcptools.com/api/mcp

Stateless, read-only, no auth, no API key. This is also the entry published to the official MCP Registry as io.github.shibley/mymcptools.

Local stdio adapter

The code in this repo is a thin stdio adapter over that same endpoint, for clients that don't speak Streamable HTTP yet. It is not on npm yet, so build it from source (Node.js 18+):

bash
git clone https://github.com/shibley/mymcptools-mcp-server.gitcd mymcptools-mcp-servernpm install && npm run build

Then point your client at the built entrypoint:

json
{  "mcpServers": {    "mymcptools": {      "command": "node",      "args": ["/absolute/path/to/mymcptools-mcp-server/dist/index.js"]    }  }}

Add that to claude_desktop_config.json (Claude Desktop), .cursor/mcp.json (Cursor), or your client's equivalent.

Set MYMCPTOOLS_MCP_URL to override the upstream endpoint the adapter talks to.

Tools

ToolWhat it answers
search_mcp_serversWhich servers match this keyword, category, client integration or install type?
get_mcp_serverFull entry for one server: install command, repo, supported clients, current probe verdict.
get_server_statusIs this server reachable right now? Verdict, exposed tool count, handshake latency, negotiated protocol version.
get_server_historyTrailing probe history and a daily uptime sparkline for one server.
list_server_incidentsReconstructed outages — contiguous runs of failed probes, with start, end and duration.
list_schema_driftWhich servers added, removed or changed tools between probes. Useful for spotting breaking changes.
list_categoriesEvery category and client integration, with counts. Returns the valid slugs for the filters above.
get_catalog_statsPopulation-level health: verdict breakdown, transport mix, latency percentiles.

What the probe data does and does not cover

Worth stating plainly, because the numbers are easy to misread:

  • The catalog carries ~2,700 entries, but only ~44 are probeable — that is, they publish a remote endpoint something can actually connect to.
  • The rest are local-only servers (stdio, run via npx or docker on your machine). There is no remote endpoint to handshake against, so they are reported as UNPROBEABLE rather than assigned a made-up status. For those, get_mcp_server falls back to a static repo-freshness signal.
  • UNPROBEABLE therefore means "not remotely checkable", not "broken".

So treat the live-verification tools as high-confidence for remote/hosted servers, and treat the catalog entry itself as the useful part for local ones.

Verdicts

VerdictMeaning
GOODHandshake succeeded and the server listed its tools.
WARNReachable, but something was off — slow handshake or a degraded response.
AUTH_REQUIREDEndpoint is live but needs credentials before it will list tools.
DOWNEndpoint is published but the handshake failed.
UNPROBEABLENo remote endpoint to probe — typically a local stdio server.

Development

bash
npm installnpm run build

Smoke-test the built server over stdio:

bash
printf '%s\n%s\n%s\n' \  '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"t","version":"1"}}}' \  '{"jsonrpc":"2.0","method":"notifications/initialized"}' \  '{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}' \  | node dist/index.js

License

MIT

来源：README.md，提交 9159ce4

工具

8

search_mcp_servers

Search MCP servers

Search the MyMCPTools catalog of Model Context Protocol servers. Filter by free-text query, category slug, integration slug (claude-desktop, cursor, vs-code, ...), install type, or official status. Set only_verified to restrict to servers that answered a live MCP handshake on the most recent probe. Results carry the live verdict for each server.

输入参数limitquerycategoryofficialintegrationinstall_typeonly_verified

get_mcp_server

Get one MCP server

Full catalog entry for one MCP server by slug: description, author, install command, repo, supported clients, current probe verdict, static repo-freshness signal for local servers, and related servers in the same categories.

输入参数slug必填

get_server_status

Get live MCP server status

Current live-probe status for one server: verdict (GOOD / WARN / AUTH_REQUIRED / DOWN / UNPROBEABLE), exposed tool count, handshake latency, negotiated protocol version, remote endpoint, and when it was last seen healthy. Omit the slug to get the catalog-wide verdict summary instead.

输入参数slug

get_server_history

Get MCP server uptime history

Trailing probe history for one server plus a daily uptime sparkline — the reliability signal behind the badge on mymcptools.com. Returns per-day uptime buckets and the raw probe points (timestamp, verdict, latency).

输入参数daysslug必填limit

list_server_incidents

List MCP server outage incidents

Reconstructed outage incidents (contiguous runs of failed probes) across the catalog, newest first. Each incident has a start, an end (or ongoing), a duration and the failure reason. Filter to one server with slug, or to open outages with status='ongoing'.

输入参数sluglimitstatus

list_schema_drift

List MCP tool-schema drift

Tool-schema and protocol-version drift events detected between successive probes — which servers added, removed or changed tools, and when. Use this to spot breaking changes in a server you depend on. Newest first.

输入参数sluglimit

list_categories

List catalog categories

Every category and client integration in the MyMCPTools catalog, with the number of servers in each. The slugs returned here are the valid category and integration filter values for catalog searches.

无需输入参数。

get_catalog_stats

Get catalog-wide health stats

Aggregate health of the whole MCP population: verdict breakdown, share of probeable servers actually serving, transport mix, handshake latency percentiles, tool counts and probe freshness. This is the 'how healthy is MCP right now?' headline number.

无需输入参数。

版本历史

1
  1. v0.1.1最新Sep 16, 2026

评分

已满足适用检查项的 100%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分