@applyra/mcp-server

中文文档
1.8.0发布于 2026/10/27MITES Module近 30 天下载 2432

MCP server for Applyra: connect your App Store and Google Play keyword data to AI assistants

原始使用文档
文档目录12 节
  1. Remote server (ChatGPT, Claude, Gemini CLI, VS Code, Cursor)
  2. Prerequisites (local server)
  3. Installation
  4. Claude Desktop
  5. Cursor
  6. VS Code (GitHub Copilot)
  7. Claude Code
  8. Codex
  9. Windsurf
  10. Available Tools
  11. Learn more
  12. License

@applyra/mcp-server

npm CI License: MIT

MCP (Model Context Protocol) server for Applyra. It connects your App Store and Google Play keyword data to AI assistants like ChatGPT, Claude, Cursor, Codex, VS Code Copilot, and more.

30 tools covering keyword rank tracking, difficulty and traffic scoring, listing audits and metadata simulation, competitor visibility, notes on your charts and what followed each one, autocomplete mining, niche clustering, and top charts, on the App Store and Google Play.

Remote server (ChatGPT, Claude, Gemini CLI, VS Code, Cursor)

The same tools are hosted at https://www.applyra.io/api/mcp (Streamable HTTP, OAuth). Nothing to install and no API key: the client opens an Applyra sign-in page, you approve it, and it is connected. It works with any Applyra account: the free plan includes 50 requests a month (each tool call is one request) on 1 app, 5 keywords and 1 competitor. Plan details are at applyra.io/pricing.

  • ChatGPT: on chatgpt.com, Plugins > Add > Create custom MCP server, with this URL and OAuth authentication, then Create as a plugin. Read and write tools both work on a ChatGPT Plus plan.
  • Claude: add it from the Claude directory in one click (on the web, desktop, mobile and in Claude Code), or as a custom connector with this URL (Customize > Connectors > Add custom connector).
  • Gemini CLI: gemini extensions install https://github.com/applyra-io/mcp-server, then sign in when Gemini CLI asks.
  • VS Code, Cursor, Perplexity, Smithery and any client that supports remote MCP servers with OAuth: add this URL as a remote (HTTP) server.

Connected assistants are listed, and can be disconnected, at applyra.io/dashboard/mcp.

Prerequisites (local server)

Installation

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "applyra": {
      "command": "npx",
      "args": ["-y", "@applyra/mcp-server"],
      "env": {
        "APPLYRA_API_KEY": "your_api_key"
      }
    }
  }
}

Cursor

Install the Applyra plugin from the Cursor Marketplace and set the APPLYRA_API_KEY environment variable. The plugin also ships an ASO research skill that guides the agent through the tools.

Or add the server manually to .cursor/mcp.json or ~/.cursor/mcp.json:

{
  "mcpServers": {
    "applyra": {
      "command": "npx",
      "args": ["-y", "@applyra/mcp-server"],
      "env": {
        "APPLYRA_API_KEY": "your_api_key"
      }
    }
  }
}

VS Code (GitHub Copilot)

Add to .vscode/mcp.json:

{
  "servers": {
    "applyra": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@applyra/mcp-server"],
      "env": {
        "APPLYRA_API_KEY": "your_api_key"
      }
    }
  }
}

Claude Code

claude mcp add applyra -e APPLYRA_API_KEY=your_api_key -- npx -y @applyra/mcp-server

Codex

Note --env, where Claude Code takes -e.

codex mcp add applyra --env APPLYRA_API_KEY=your_api_key -- npx -y @applyra/mcp-server

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "applyra": {
      "command": "npx",
      "args": ["-y", "@applyra/mcp-server"],
      "env": {
        "APPLYRA_API_KEY": "your_api_key"
      }
    }
  }
}

Available Tools

ToolDescription
list_applicationsList tracked apps with store metadata, ratings, keyword count and ASO Health scores
add_applicationTrack a new application by its store bundle ID. Fetches store metadata and computes the initial visibility score
list_keywordsTracked keywords with current rank, favorite flag, difficulty/traffic scores
track_keywordsTrack up to 20 new keywords for an application in a single call
untrack_keywordStop tracking a keyword for an application (soft delete)
set_keyword_favoriteMark or unmark a tracked keyword as favorite for a specific app
inspect_keywordDeep-analyze any keyword: difficulty, traffic, KEI, top 20 apps, related keywords
list_keyword_inspectionsPast keyword inspections with their scores
run_autocompleteFetch autocomplete suggestions from the App Store or Google Play
list_autocomplete_historyPast autocomplete queries
run_niche_analysisCluster a niche topic into sub-niches with opportunity scores
list_niche_analysesPast niche analyses
top_chartsTop apps chart for a store/country/category/collection, with daily rank movement
list_top_chart_categoriesCategories and collections supported by top_charts, per store
get_keyword_rank_historyDaily rank evolution over a date range
get_app_score_historyDaily visibility score history for an app, with the notes and updates in the range
list_notesThe notes on an app's charts, numbered like their flags
add_noteNote what you did on a day (a campaign, a metadata change, a launch)
update_noteChange the text or the day of a note
delete_noteDelete a note
compare_eventsVisibility, or a keyword's rank, before and after each note and app update
get_aso_healthFull listing audit for an app: coverage, targeting and appeal, every field, and what to fix
check_metadataCheck draft listing text against each store's limits and forbidden copy. Instant, no store lookup
simulate_metadataScore a listing that does not exist yet and see the gain against the app's current score
list_metadata_simulationsListing drafts already scored on the account (titles and scores)
get_metadata_simulationOne saved draft in full: its four fields, its context and its findings
list_competitorsCompetitor pairs with side-by-side visibility scores
add_competitorAdd a competitor app to one of your applications by its store bundle ID
remove_competitorRemove a competitor relationship by its internal ID
get_account_usageCurrent usage vs. plan limits

Learn more

License

MIT. See LICENSE.

版本与发布记录

@applyra/mcp-server 共发布 14 个版本,首次发布于 2026/03/07;过去 12 个月发布了 14 个版本。

版本发布日期距今
1.8.0latest2026/10/2711 天前
1.7.02026/10/1326 天前
1.6.22026/10/071 个月前
1.6.12026/10/071 个月前
1.6.02026/10/071 个月前
1.5.02026/09/171 个月前

关于 @applyra/mcp-server 的常见问题

@applyra/mcp-server 是什么?

MCP server for Applyra: connect your App Store and Google Play keyword data to AI assistants

如何安装 @applyra/mcp-server?

在项目目录运行 npm install @applyra/mcp-server,或使用 pnpm add / yarn add / bun add。它还提供命令行工具 mcp-server,可以用 npx mcp-server 直接运行。

@applyra/mcp-server 支持 TypeScript 吗?

该包的 package.json 没有声明类型入口。需要类型提示时,可以在 npm 上查找社区维护的 @types/applyra__mcp-server 是否存在。

@applyra/mcp-server 支持 ESM 还是 CommonJS?

以 ES Module 形式发布,推荐使用 import 引入。在 CommonJS 项目中,Node.js 20.19+ / 22.12+ 可以直接 require() 不含顶层 await 的 ES Module,更早的版本需要改用动态 import()。

@applyra/mcp-server 需要什么版本的 Node.js?

package.json 的 engines 字段要求 Node.js >=20。版本不符时 npm 默认只给出警告,开启 engine-strict 后会拒绝安装。

@applyra/mcp-server 可以商用吗?

@applyra/mcp-server 使用 MIT 许可。宽松许可:允许商业使用、修改和再分发,通常只需在分发时保留原版权与许可声明。以上为概括,不构成法律意见,请以许可原文为准。

@applyra/mcp-server 的下载量有多少?

根据 npm 官方下载统计,@applyra/mcp-server 在 2026/10/07 – 2026/11/05 这 30 天内被下载 2,432 次,日均约 81 次。下载量会把 CI 构建、镜像同步和作为其他包依赖的安装一并计入,适合横向比较热度,不能直接等同于使用人数。

@applyra/mcp-server 还在维护吗?

最新版本 1.8.0 发布于 2026/10/27(11 天前);至今共发布 14 个版本,其中过去 12 个月发布了 14 个;首次发布于 2026/03/07。发布频率只是参考信号:功能稳定的工具可能很少需要更新,建议结合源码仓库的 issue 和提交记录判断。