GitHub
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
作者提供的说明(英文原文)
Zones, DNS, cache purge, Workers, KV, and R2 through the Cloudflare API.
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": {
"edgeport": {
"command": "npx",
"args": [
"-y",
"@reneturcios/edgeport",
"mcp"
]
}
}
}在终端中运行这条命令
claude mcp add edgeport -- npx -y @reneturcios/edgeport mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"edgeport": {
"command": "npx",
"args": [
"-y",
"@reneturcios/edgeport",
"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": {
"edgeport": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/edgeport",
"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7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5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 Node.js, Docker。
同一分类中有:GitHub, Desktop Commander, Cloudflare MCP
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
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
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。
edgeport is a small doorway into a Cloudflare account. From one bin an agent can see which accounts and zones a token can reach, read and change DNS records, purge cached content, and inventory the account's Worker scripts, KV namespaces, and R2 buckets. It runs from the command line or as an MCP server for agents.
A Cloudflare API token is required. Put it in the CLOUDFLARE_API_TOKEN environment variable; it is sent to the API as a Bearer token and never appears in output. Account-scoped tools (Worker scripts, KV, R2) take an account_id argument, and when that is omitted they fall back to CLOUDFLARE_ACCOUNT_ID. When a required variable is missing, the call fails with an error that names it.
This is a core subset of the Cloudflare API, not the whole surface: Workers bindings deep configuration, observability and Logpush, rulesets and firewall rules, Pages and deployments, Stream and Images, AI and Vectorize, and the hosted documentation search tool are out of scope.
npx @reneturcios/edgeport list-zones
Or run the container image:
docker run -i -e CLOUDFLARE_API_TOKEN ghcr.io/cloudcorprecords/edgeport
The container defaults to the MCP server (see below). Pass CLI arguments to use it as a CLI instead.
export CLOUDFLARE_API_TOKEN="your api token"
export CLOUDFLARE_ACCOUNT_ID="your account id" # optional fallback
edgeport list-accounts edgeport list-zones --name example.com edgeport get-zone --zone-id edgeport list-dns --zone-id --type A edgeport purge-cache --zone-id --everything edgeport list-workers edgeport list-kv edgeport list-r2
edgeport --version
Exit codes: 0 on success, 1 when an API call fails (including a missing CLOUDFLARE_API_TOKEN or a success: false envelope), 2 for usage errors.
One bin serves both faces. The mcp subcommand starts a stdio MCP server:
npx -y @reneturcios/edgeport mcp
MCP client configuration:
{
"mcpServers": { "edgeport": { "command": "npx", "args": ["-y", "@reneturcios/edgeport", "mcp"], "env": { "CLOUDFLARE_API_TOKEN": "your api token" } } } }
| Tool | Input | What it returns |
|---|---|---|
list_accounts | per_page (integer, optional) | Accounts the token can access, from the envelope's result. |
list_zones | account_id, name, per_page (all optional) | Zones, optionally filtered by account or name. |
get_zone | zone_id (required) | One zone's details. |
list_dns_records | zone_id (required), type, name, per_page (optional) | DNS records in the zone. |
create_dns_record | zone_id, type, name, content (required), ttl, proxied, comment (optional) | The created record. |
update_dns_record | zone_id, record_id (required), any record field (optional) | The updated record; only the fields given are sent. |
delete_dns_record | zone_id, record_id (required) | The delete result. |
purge_cache | zone_id (required), purge_everything, files, tags, hosts (optional) | The purge result for the zone. |
list_workers_scripts | account_id (optional) | Worker scripts in the account. |
get_worker_script | script_name (required), account_id (optional) | A shape summary only: content type, byte size, body present. The script body is never returned. |
list_kv_namespaces | account_id (optional) | Workers KV namespaces in the account. |
list_r2_buckets | account_id (optional) | R2 buckets in the account. |
Every JSON response is unwrapped from Cloudflare's {success, errors, result} envelope. A success: false envelope is reported as a tool error carrying the API's own error entries.
npm install
npm run build # tsc -> dist/ npm test # node test runner over test/*.test.ts, no network (fake fetch fixtures) npm run dev # run the CLI from source via tsx
MIT — see LICENSE.