
HappyKeywords
io.github.yurybreiapps-jpg由 GitHub 账号 yurybreiapps-jpg 发布v1.0.0更新于 Nov 13, 2026
App Store keyword research (ASO): ranks in 50+ countries, demand, difficulty. Needs the Mac app.
概览
通过 Mac 应用进行 App Store 关键词研究(ASO):50 多个国家/地区的排名、关键词需求与难度、商店页审核与本地化。
- 可以做什么
- 跟踪自己与竞品应用在 50 多个国家/地区的每日排名
- 查询关键词的预估需求与难度
- 运行商店页 Audit 并获得按优先级排序的改进建议
- 撰写并本地化名称、副标题与关键词方案
- 将商店页改动上传到 App Store Connect 的草稿版本
- 发掘热门细分市场与产品创意,长任务通过 wait 跟进
- 适用场景
- 适合推广或开发 iOS 应用的人,希望让助手在对话中挑选关键词、比较各国排名、审核商店页或起草本地化元数据时使用。
- 运行要求
- 需要 macOS 14 或更高版本(Apple 芯片或 Intel),并在 /Applications 中安装 HappyKeywords 应用;MCP 服务器由该应用以本地 stdio 进程启动。调用工具需要 7 天试用或一次性许可证。App Store Connect 访问权限在应用内配置。
安装
在 SourceWeft 中
- 打开 控制台中的 HappyKeywords,将其添加到工作区。
- 为需要使用其工具的对话启用该服务。
Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。
其他 MCP 客户端
参照 仓库 中的启动说明。
README
HappyKeywords MCP server
App Store keyword research (ASO) from your AI agent. Ask Claude Code, Codex, Cursor, Claude or any MCP app which keywords to target, where your app ranks in 50+ countries, and what to fix in your listing.
The MCP server is built into HappyKeywords, a Mac app. This repo holds the setup and the tool list, plus a small bundle that starts the app's server; the app's source isn't here. Docs: https://happykeywords.com/docs/mcp
Watch: a one-minute demo
[Claude Code asks HappyKeywords which App Store keyword to go after next]
Claude Code reads an app's real ranks, demand and difficulty from HappyKeywords and picks the keyword to go after next. Watch on YouTube · Vertical version
What it does
- Track your apps and competitors in 50+ countries, with daily ranks
- Estimated demand and difficulty for every keyword (0–100 estimates from App Store search results, not Apple's search counts)
- An Audit that scores your listing out of 100 and says what to fix first
- Write and localize name, subtitle and keyword options
- Upload them into your draft version in App Store Connect: a preview first, Undo after. Nothing is ever submitted for review.
- Hot niches and product ideas before you build
112 tools. Long jobs run inside the app; the agent follows them with wait.
Requirements
- macOS 14 or later (Apple silicon or Intel)
- HappyKeywords in /Applications: download at https://happykeywords.com/?ref=github-mcp
- 7-day free trial, no card. Then a one-time license ($29 in Early Access). No subscription.
- Tools list without a license; tool calls need the trial or a license.
Connect
One click: in HappyKeywords, open Settings → AI apps (MCP) and press Add next to your agent.
Claude desktop, as a bundle: download happykeywords.mcpb from the
latest release and open it.
The bundle only starts the MCP server inside the installed app (source in bundle/).
Or by hand:
Claude Code
Codex
Cursor, Claude desktop and other MCP apps (~/.cursor/mcp.json, claude_desktop_config.json)
Try asking
- Which of my apps lost ranks since the last check?
- Where is my app outside the top 10 in Germany, and what would it take to get in?
- Run the Audit for my app and tell me what to fix first.
- Write three name and subtitle options, then localize the best one into German and French.
- Find hot niches in the US App Store and show me product ideas for the best one.
Tools
Full list with inputs: https://happykeywords.com/docs/mcp#tools
Privacy
Everything runs on your Mac. App Store Connect keys stay in the macOS Keychain, and no MCP tool can read them. Uploads happen only after you approve the preview.
Support
来源:README.md,提交 35c5807
工具
0版本历史
1- v1.0.0最新Nov 13, 2026
评分
已满足适用检查项的 40%
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25 分
请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。
- 已声明认证方式10 分
请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。
- 已列出工具15 分
请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。
- 描述必需15 分
- 图标10 分
- 文档15 分
- 许可证10 分
请在仓库或注册中心条目中声明许可证。