Prototype · Preview: every server in the official registry, synced every 30 minutes. Reviews and accounts come in a later step.
PopNx MCPMCP servers for people who use AI · explained in plain words

FeedWatch

by bruceonline2026version 0.1.0i★ –iRuns in the cloud
Registry nameio.github.bruceonline2026/feedwatch

Install

Overview

Description from the maker (original English)

Monitor RSS/Atom/JSON feeds; alerts on new items AND edits to existing ones, with field-level diffs.

PopNx MCP’s three pillars

These three are never merged into one score: opinions, measurements and popularity are different things.

What you can do with it

We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.

  • Connection: Runs in the cloud
  • Transport: Streamable HTTP
  • Category: Developer tools

Install

Pick the AI app you use, then copy the settings.

Before you install

  • Basic checks: 1 of 2 passed Fail
  • Runs on the maker’s server: what you ask it goes through their server.
  • The registry lists no key to fill in.

You need

Which app do you use?

Pick your app above. We remember it, so next time the button says “Install in …”.

Works in Claude on the web, desktop and mobile:

  1. Open Settings → Connectors
  2. Press Add custom connector
  3. Paste the address below and press Add
  4. If it asks you to sign in, sign in with the maker
SettingsConnectors1+ Add2https://…3
https://api.tiici.com/mcp

Run this in your terminal

claude mcp add --transport http feedwatch https://api.tiici.com/mcp

Open in Cursor

Or copy the settings yourself

Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json

{
  "mcpServers": {
    "feedwatch": {
      "url": "https://api.tiici.com/mcp"
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

Open in VS Code

Or copy the settings yourself

Paste into .vscode/mcp.json in your project

{
  "servers": {
    "feedwatch": {
      "type": "http",
      "url": "https://api.tiici.com/mcp"
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

ChatGPT connects only to servers on the internet:

  1. Open ChatGPT on the web → Settings → Connectors (Apps)
  2. Press Create or Add (some plans need Developer mode under Advanced first)
  3. Paste the address below, give it a name and save
  4. If it asks you to sign in, sign in with the maker
SettingsConnectors1+ Create2https://…3
https://api.tiici.com/mcp

Menu names can differ slightly by plan and over time.

Your choice stays on this device in a cookie; no account needed.

These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.

What it can do

The server did not answer when we last checked, so we could not read its tools.

Basic checksi

We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 8, 2026

  • ✓ Pass

    Maker matches the repo

    The registry name was verified with the GitHub account bruceonline2026, which owns the repo

  • … Not checked yet

    Still maintained

    We have not read this repo from GitHub yet

  • … Not checked yet

    Has a licence

    We have not read this repo from GitHub yet

  • ✕ Fail

    Endpoint responds

    Did not respond within 6 seconds when we checked

  • – Not applicable

    Sign-in with OAuth

    Did not respond within 6 seconds when we checked

  • … Not checked yet

    No hidden instructions in tool descriptions

    Not checked yet; we read the tool list when we check the endpoint

Compared with others in Developer toolsi

  • More stars than 0% of the other 4907 servers in this category
  • Published to the registry more recently than 64% of the other 4907 servers in this category
  • Basic checks equal to or better than 16% of the other 4907 servers in this category

Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.

Common questions

Is FeedWatch safe?

It passed 1 of 2 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.

What keys or codes does it need?

The registry does not list a required key, though some servers ask you to sign in on first use.

Does it run on my computer or in the cloud?

It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.

Is it still maintained?

The GitHub repo last changed –, and the latest version (0.1.0) reached the registry 22 days ago.

Which apps can use it?

Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.

What can I use instead?

In the same category: GitHub, Desktop Commander, Cloudflare MCP

Alternatives

Servers in the same category from other makers, sorted by stars.

Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi

GitHub

by github

GitHub’s official server lets AI read and manage code, issues and pull requests in your account. Use GitHub’s hosted version or run it locally with Docker.

Computer or cloud★ 34,245MITUpdated 4 days ago
Install

Desktop Commander

by wonderwhy-er

MCP server for terminal commands, file operations, and process management

Runs on your computer★ 10,225MITUpdated 16 days ago
Install

Cloudflare MCP

by cloudflare

Cloudflare MCP servers

Runs in the cloud★ 4,446Apache-2.0Updated 14 months ago
Install

README from the maker

Open the README

Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.

FeedWatch MCP

RSS/Atom 变更监控 API + MCP Server 当前阶段:验证期准备(判定线:8 周内 5 个真实付费用户)

🌐 English version: README.en.md · 中文在下方


从这里开始

你想了解看这份
⭐ 现在什么状态、下一步做什么📋 工作计划与进展
整个项目长什么样📐 项目蓝图
开发要做什么💻 软件开发需求说明书
客户怎么用它📖 用户说明书
买服务器 + 部署🖥️ 服务器采购与部署手册
注册海外公司🏢 海外公司注册与维护手册
给客户签发API Key[🔑 API Key 管理与配额](docs/运营手册/API Key管理与配额.md)
自己先试用一遍🧪 内部测试指南
发布到MCP 生态🔌 MCP生态发布指南
收款开通(Payoneer)💳 Payoneer 收款链接操作手册
收款怎么配(通用)💰 收款与运营手册
为什么这么选📋 决策记录-收款与主体
改完代码要跑什么🧪 回归测试要求点清单
提交前自查(4 问 + 16 项检查)✅ 提交前自查清单
踩过什么坑🔍 调试复盘
上线要查什么✅ 上线检查清单
状态字段什么意思🔢 状态字段语义

⚠️ 改代码前先看「回归测试要求点清单」 —— 里面有 89 项离线测试的覆盖现状、

19 项已知缺口、以及已证伪的 4 处错误「已验证」。

不要只看测试名声称覆盖了什么,要看它实际断言了什么。


一句话说明

客户把 RSS/Atom/JSON Feed 的网址交给我们,我们替他们盯着;内容变了就主动推送, 并精确告诉他哪个字段、从什么变成什么。

核心差异:竞品多数只能报"新增条目",我们能检测已有条目被修改并给出字段级 diff—— 新闻更正错字、竞品悄悄改价格,这些都不产生新条目,但恰恰是客户最需要知道的。


文档体系(四层结构)

docs/

├── 架构/ 【第 1 层】顶层设计 │ └── 项目蓝图.md 整个项目 + 架构 + 外部资源 + 阶段路线 │ ├── 需求/ 【第 2 层】做什么 │ ├── 软件开发需求说明书.md功能清单 + 验收标准 + 需求来源 │ └── 用户说明书.md 客户视角的产品说明与使用场景 │ ├── 运营手册/ 【第 3 层】怎么操作 │ ├── 服务器采购与部署手册.md 选型对比 + Vultr 注册购买 + Linux 部署 │ ├── 海外公司注册与维护手册.md 香港公司注册 + 银行开户 + 后续维护 │ ├── 内部测试指南.md ⭐ 8 步实测流程(含MCP 与安全验证) │ ├── API Key管理与配额.md ⭐ 签发 Key + 各项限流 + 配额口径核对 │ ├── MCP生态发布指南.md ⭐ 官方注册表 + 5 个目录收录方式 │ ├── Payoneer收款链接操作手册.md ⭐ 验证期收款开通(注意:非 Checkout) │ ├── 收款与运营手册.md 付费时机 + 收款配置 + 排障(第三部分已废弃) │ └── 部署配置与待办-Windows方案.md (已弃用,保留参考) │ ├── 决策/ 【第 4 层】为什么这么选 │ └── 决策记录-收款与主体.md 平台选型 + 主体时间线 + 被排除项(含 v2 纠错) │ └── 复盘/ 【第 5 层】经验沉淀 ├── 回归测试要求点清单.md ⭐ 89 项测试覆盖现状 + 19 项缺口 + 最小回归集 ├── 调试复盘.md 10 个缺陷 + 5 条交付规范 + 失误归因 ├── 上线检查清单.md 三阶段 checklist + B1–B6 ├── 状态字段语义.md 三状态正交表 + 错误码判读 └── 事故复盘/ └── 2026-10-13-www站点500事故复盘.md

分层判断标准
层回答什么问题变更频率
架构整个项目长什么样很少(季度级)
需求做什么功能、验收标准中(月级)
运营手册怎么一步步做高(跟着平台变化)
决策为什么这么选、为什么排除别的低(决策后很少变)
复盘踩过什么坑、教训是什么追加式

⚠️ 常见错误:把"为什么选 A 不选 B"写进运营手册。 平台会变(被收购、改费率),手册不会自动更新 —— 一份三个月前正确的手册,今天可能把人导向一个用不了的平台。


快速开始(开发者)

npm install

HOST=127.0.0.1 PORT=3001 PUBLIC_URL=https://api.tiici.com \ FEEDWATCH_BOOTSTRAP_KEY=<至少32位随机> \ npm start curl http://127.0.0.1:3001/health

⚠️ Bind to 127.0.0.1, never 0.0.0.0. Put Nginx (or any reverse proxy) in

front so TLS, access logs and WAF are not bypassed.

测试
npm test              # 87 项离线测试(5 个套件)

npm run test:live # 28 项真实源测试(需出网)


技术栈

运行时      Node.js 22 ESM

依赖 fast-xml-parser(唯一运行时依赖) HTTP node:http 原生(零框架) 部署 Linux + Nginx + systemd / Windows + IIS 测试 自建轻量断言框架(无 jest/vitest) 规模 2143 行 / 7 模块 测试规模 87 项离线全绿,全部经「回退法证伪」

为什么零框架依赖:便于跨环境(Linux/Windows)部署,减少供应链风险。


目录结构

06MCP/

├── web/index.html 对外落地页(部署到 /) ├── docs/ │ ├── 工作计划与进展.md ⭐ 当前状态 + 每日台账 + 下一步 │ ├── 架构/项目蓝图.md 整体架构 + 外部资源 + 阶段路线 │ ├── 需求/ 功能清单 + 实现状态 + 客户说明 │ ├── 运营手册/ ⭐ 内部测试 / Key 管理 / MCP 发布 / 收款 / 部署 │ ├── 决策/ 收款平台与主体决策(v2 含纠错) │ └── 复盘/ ⭐ 提交前自查清单 / 回归测试要求点 / 调试复盘 / 上线清单 / 状态语义 ├── src/ 7 模块 2242 行 ├── scripts/ ~2100 行 / 7 套件 │ ├── selfcheck-silent.js ★ 三个高发模式的 12 项自动检查(npm test 第一道) │ ├── selftest / regression*.js 92 项离线断言 │ ├── e2e.js 端到端(含回退证伪用例) │ └── livefeed-test.js 28 项真实源测试(需出网) ├── deploy/ Linux/Windows/落地页/云防火墙部署脚本 ├── server.json MCP 官方注册表清单 ├── smithery.yaml Smithery 一键安装配置 ├── web/index.html 对外落地页 └── deliverables/gstack/ 审查与发布报告(不随仓库公开)


当前状态

📋 完整台账见 [`docs/工作计划与进展.md`](docs/工作计划与进展.md)(每日工作 + 下一步 8 项)

✅ 已就绪
产品        https://api.tiici.com/ 落地页 + REST API + MCP 端点,全部线上

源码 2242 行 / 7 模块(零 Web 框架) 测试 16 项模式自查+ 140 项断言全绿 + 28 项真实源 = 2000 行 / 7 套件 安全 15 个缺陷全修(每个都做回退证伪)+ 三层防护 云防火墙 / ufw / Nginx+TLS,3001 公网实测不可达 服务器 Vultr 167.179.112.40(Tokyo / $5/月 / Ubuntu 24.04.5) 运行 feedwatch.service active+enabled,Node v22.23.3,实测 RSS 93 MB 证书 api.tiici.com 已签发(至 2027-01-13,自动续期 dry-run 已验证) 配额 免费层 5 个监控点 / 6h 频率(申报与实际强制一致) 分发 GitHub 公开仓库 + tag v0.1.0 + server.json 通过官方 validate 依赖 fast-xml-parser 5.11.2(OSV 漏洞清零) 文档 15 份(含回归测试要求点清单、API Key 管理、内部测试指南)

⏳ 待办(完整版见工作计划)
#事项负责阻塞什么
1Payoneer 个人账户 + 收款链接(⚠️ 非 Checkout)徐总🔴 收款前置条件
2提交 MCP 官方注册表徐总分发(mcp-publisher publish)
3点 GitHub Release Publish徐总tag 已推送,只需点一下
4Smithery / mcp.so / mcpservers.org徐总各 3-5 分钟
6咨询税务师(第 3 周)徐总后续合规
7TTL 调回 300徐总—

⏸️暂缓:海外公司注册 —— 验证期不需要,确定要接 Checkout 2.0 / Stripe 时再办

📌 不在本项目范围:www.tiici.com / game.tiici.com 的证书续签属于

WordMaster 等其他站点,归各自项目台账管理,本台账不跟进。


核心判定线

指标目标
真实付费用户≥ 5 个(8 周内)
推送准确率/漏报率< 1%
转化率(免费→付费)> 5%

⚠️ 不要用"技术跑通"当判定 —— 要有人掏钱。技术跑通只是必要条件。


⚠️ 两条给后续接手者的提醒

① 不要只看文档里的「已验证」三个字

本次审查里,团队成员至少 5 次写了「已验证」而结论后来被推翻。 可靠的方式是:临时把修复改回去,看测试会不会红。

② 声称「X 会失败」前,先证明 X 在应该成功时确实成功

只测失败路径无法区分「X 有问题」和「工具有问题」。


遇到问题先看 `docs/复盘/调试复盘.md` —— 那里记录了 10 个真实缺陷与它们被发现的经过,

避免重复踩坑。

完整审查报告:deliverables/gstack/pre-launch-review-feedwatch-2026-10-12.md

Ask PopNx