能力标签
🔌
MCP工具

开源MCP工具:捕获iOS和macOS应用截图

基于 Swift · 让 AI 助手直接操作你的系统与工具
英文名:storescreens-cli
⭐ 19 Stars 🍴 1 Forks 💻 Swift 📄 MIT 🏷 AI 7.5分
7.5AI 综合评分
app-storecliiosmacosswift
✦ AI Skill Hub 推荐

经 AI Skill Hub 精选评估,开源MCP工具:捕获iOS和macOS应用截图 获评「推荐使用」。这款MCP工具在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 7.5 分,适合有一定技术背景的用户使用。

📚 深度解析
开源MCP工具:捕获iOS和macOS应用截图 是一款基于 MCP(Model Context Protocol)标准协议的 AI 工具扩展。MCP 协议由 Anthropic 开发并开源,旨在建立 AI 模型与外部工具之间的标准化通信接口,目前已被 Claude Desktop、Claude Code、Cursor 等主流 AI 工具采纳。

通过安装 开源MCP工具:捕获iOS和macOS应用截图,你的 AI 助手将获得额外的工具调用能力,可以用自然语言直接操控该工具的功能,无需学习复杂的命令行语法。MCP 工具的核心价值在于"一次配置,永久增强"——配置完成后,每次与 AI 对话时都可以无缝调用这些工具。

在技术实现上,MCP 工具通过标准的 JSON-RPC 协议与 AI 客户端通信,工具的功能以"工具列表"的形式暴露给 AI 模型,AI 可以按需调用。开源MCP工具:捕获iOS和macOS应用截图 提供了结构化的工具调用接口,使 AI 模型能够精确地理解和使用每个功能点,显著降低 AI 在工具使用上的错误率。

与传统的 API 集成相比,MCP 工具的优势在于无需编写代码——用户只需在配置文件中添加几行 JSON,即可让 AI 获得全新能力。AI Skill Hub 将 开源MCP工具:捕获iOS和macOS应用截图 评为 AI 评分 7.5 分,属于同类工具中的优质选择。
📋 工具概览

开源MCP工具:捕获iOS和macOS应用截图 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。

GitHub Stars
⭐ 19
开发语言
Swift
支持平台
macOS / iOS
维护状态
轻量级项目,按需更新
开源协议
MIT
AI 综合评分
7.5 分
工具类型
MCP工具
Forks
1
📖 中文文档
以下内容由 AI Skill Hub 根据项目信息自动整理,如需查看完整原始文档请访问底部「原始来源」。

开源MCP工具:捕获iOS和macOS应用截图 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。

📌 核心特色
  • 通过标准 MCP 协议与 Claude、Cursor 等主流 AI 客户端深度集成
  • 提供结构化工具调用接口,显著降低 AI 集成复杂度
  • 支持 Claude Desktop 和 Claude Code 无缝接入,开箱即用
  • 可与其他 MCP 工具组合叠加,构建完整 AI 工作站
  • 轻量无侵入设计,不影响现有系统架构
🎯 主要使用场景
  • 在 Claude Desktop 对话中直接调用本地工具,实现 AI 与系统的深度联动
  • 通过自然语言驱动复杂的多步骤自动化任务,代替繁琐手动操作
  • 将多个 MCP 工具组合使用,构建个人专属 AI 工作站
以下安装命令基于项目开发语言和类型自动生成,实际以官方 README 为准。
安装命令
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/ciscoriordan/storescreens-cli

# 方式二:手动配置 claude_desktop_config.json
{
  "mcpServers": {
    "--mcp-----ios-macos----": {
      "command": "npx",
      "args": ["-y", "storescreens-cli"]
    }
  }
}

# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
📋 安装步骤说明
  1. 确认已安装 Node.js(v18 或以上版本)
  2. 打开 Claude Desktop 或 Claude Code 的 MCP 配置文件
  3. 按「交给 Agent 安装 → Claude Desktop」标签中的 JSON 配置填入 mcpServers 字段
  4. 保存配置文件并重启 Claude 客户端
  5. 重启后,在对话中即可使用本工具
以下用法示例由 AI Skill Hub 整理,涵盖最常见的使用场景。
常用命令 / 代码示例
# 安装后在 Claude 对话中直接使用
# 示例:
用户: 请帮我用 开源MCP工具:捕获iOS和macOS应用截图 执行以下任务...
Claude: [自动调用 开源MCP工具:捕获iOS和macOS应用截图 MCP 工具处理请求]

# 查看可用工具列表
# 在 Claude 中输入:"列出所有可用的 MCP 工具"
以下配置示例基于典型使用场景生成,具体参数请参照官方文档调整。
配置示例
// claude_desktop_config.json 配置示例
{
  "mcpServers": {
    "__mcp_____ios_macos____": {
      "command": "npx",
      "args": ["-y", "storescreens-cli"],
      "env": {
        // "API_KEY": "your-api-key-here"
      }
    }
  }
}

// 保存后重启 Claude Desktop 生效
📑 README 深度解析 真实文档 完整度 80/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理,保留代码块、表格与列表结构。

StoreScreens

Every App Store Connect API call. Granular, agentic screenshot rendering.

StoreScreens is a brew-installed Swift MCP/CLI (and MD Skill) that drives the entire App Store Connect pipeline from one config file: XCUITest screenshots, framed renders with markdown captions (optionally unique per device/locale), metadata and binary upload via Apple's official API. Beautiful, modern bezels, no Ruby version hell.

Captures run your UI tests on multiple simulators in parallel (or natively on macOS), organize the output by device and locale, and auto-detect which App Store size each simulator maps to. Supports iPhone, iPad, Apple Watch, and Mac App Store screenshots.

Prerequisites

  1. Create an App Store Connect API key at https://appstoreconnect.apple.com/access/integrations/api. Choose either Admin or App Manager access. Download the AuthKey_XXXXXX.p8 file and keep it safe; Apple only lets you download it once.
  2. Record the Key ID (10-character alphanumeric) and Issuer ID (a UUID) from the same page.

Install

Requires macOS 14+ (Sonoma or later) on Apple Silicon (arm64). Intel Macs are not supported.

Confirm the test target builds and the test is discoverable

xcodebuild build-for-testing \ -workspace Example.xcworkspace \ -scheme Example \ -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \ 2>&1 | tee build.log

`storescreens setup`

Interactive wizard that generates a screenshot test file and wires it into your config. It scans your Swift source to auto-discover screens from TabView, NavigationLink, .sheet, .fullScreenCover, and Navigator route patterns.

$ storescreens setup

Project Detection
  ✓ Found project: MyApp.xcodeproj
  ✓ Detected scheme: MyApp

UI Test Target
  ✓ Found UI test target: MyAppUITests

Screenshot Screens
  Found 4 screens in your source code:
    1. Home (TabView)
    2. Search (TabView)
    3. Settings (NavigationLink)
    4. Profile (sheet)

  Press Enter to use these, or type your own (comma-separated)
  >
  ✓ Using discovered screens.

  ✓ Wrote MyAppUITests/ScreenshotTests.swift (4 screenshots)
  ✓ Updated storescreens.yml

If no screens are found in your source, it falls back to asking you to type them manually. If no UI test target exists, it prints step-by-step instructions to create one in Xcode.

Use --non-interactive to skip prompts and use auto-discovered screens (or defaults if none found).

(default), or "verbose" (full xcodebuild output).

builds. When unset, a per-run temp dir is used and cleaned up after.

Install

  1. Download DMGs from https://developer.apple.com/design/resources/ (Product Bezels section; iPhone, iPad, MacBook as needed).
  2. Double-click each DMG to mount.
  3. Run:
storescreens bezels import

This auto-scans /Volumes/ for Apple Design Resource DMGs, classifies PSDs by screen pixel dimensions, applies your colorway preferences, and exports transparent-screen PNGs + JSON sidecars to ~/Library/Application Support/storescreens/bezels/.

Quick Start

```bash cd /path/to/your/xcode-project

custom: https://example.com/my-flag.svg

Quick start

render:
  enabled: true
  output_dir: ./storescreens-framed

  background:
    color: "#1a1a2e"

  caption:
    title:
      font: system
      weight: bold
      font_size_pct: 5.5
      color: "#ffffff"
    min_height_pct: 22

  chrome:
    style: stroke
    stroke_color: "#ffffff"
    stroke_width: 3

  slides:
    "Home":
      caption: "Your recipes, organized."
    "Search":
      caption:
        - Find anything
        - in *seconds*.
    "Detail":
      caption:
        title: Every **detail**, at a glance.
        subtitle: Powered by AI
        highlights:
          - { match: detail, color: "#feb909", weight: heavy }

Run the render independently:

storescreens render

Or run capture with the render pass included (auto-enabled by render.enabled: true):

storescreens capture

Use --no-render to skip the render pass on a given capture run.

2. Generate screenshot UI tests

storescreens setup

4. Capture screenshots on all devices (--verbose for live output)

storescreens capture --verbose ```

`storescreens screenshot`

Takes a quick screenshot of a running simulator's current screen. No build, no tests - just captures whatever is on screen and saves it to a file. Intended for quick visual checks during UI development.

```bash

Screenshot the first booted simulator

storescreens screenshot

Screenshot a specific simulator

storescreens screenshot --simulator "iPhone 17 Pro" --output screenshot.png

screenshots whose name appears here are kept.

screenshots:

battery) for clean screenshots. Default: true. Set to false to leave

operator; override when a specific screenshot needs different values.

Rendering captioned screenshots

storescreens can post-process captured screenshots into framed, captioned images suitable for App Store Connect uploads. Add a render: block to storescreens.yml and run storescreens render (or let it run automatically after storescreens capture).

1. Generate a config file

storescreens init

Configuration

storescreens.yml:

```yaml project: "MyApp.xcodeproj" scheme: "MyApp"

devices: - simulator: "iPhone 17 Pro Max" - simulator: "iPhone 17 Pro" - simulator: "iPad Pro 13-inch (M5)" # macOS devices run tests natively (no simulator) # - simulator: "Mac 2560x1600" # platform: macOS

Multiple locales (optional) - runs the full capture once per locale

Render pass (optional): composites captioned images from captures

App Store search preview (optional): renders a faithful iPhone search

Configure credentials

Either set environment variables (CI-friendly):

export ASC_KEY_ID=ABCDE12345
export ASC_ISSUER_ID=69a6de84-03c8-47e3-e053-5b8c7c11a4d1
export ASC_KEY_PATH=~/.appstoreconnect/AuthKey_ABCDE12345.p8

Or generate a pre-filled credentials template and edit it:

storescreens auth init

This writes ~/.storescreens/asc-credentials.yml (0600 perms) with commented placeholders for key_id, issuer_id, and key_path, and opens it in your editor. Replace the three REPLACE_ME values with your real credentials.

Or run the interactive login that prompts for each value:

storescreens auth login

Either way, verify with:

storescreens auth status

This mints a JWT and hits /v1/users to confirm the key works.

Colorway / model preference

By default the importer picks "Space Black" when available, else Silver / Natural Titanium. Override per project:

render:
  chrome:
    style: bezel
    model_preference: [Pro Max, Pro, Air]
    colorway_preference: ["Cosmic Orange", Silver]

Three pieces, one workflow

StoreScreens ships as three complementary pieces. Most users only need the CLI; the other two exist to make AI coding assistants first-class operators.

PieceWhat it isWhen you want it
storescreens (CLI)The core binary. Runs UI tests across simulators, captures screenshots, builds the HTML preview gallery.Always - this is the engine. Use it from your terminal, CI, or scripts.
storescreens-mcp (MCP server)A structured wrapper that exposes the CLI's operations as [Model Context Protocol](https://modelcontextprotocol.io) tools with inline progress streaming.When your AI coding assistant (Claude Code, Cursor, etc.) should drive captures directly instead of parsing CLI output from a Bash call.
[storescreens-skill](https://github.com/ciscoriordan/storescreens-skill)An agent skill - instructions and templates that teach an assistant how to detect your Xcode project, generate config, scaffold UI tests, and run a capture.When you want an assistant to do the full setup for you, from zero to first screenshots, with no manual steps. Works with any assistant that supports skills.

Both the CLI and MCP server are installed by brew install storescreens.

How this compares to other Xcode MCP servers

StoreScreens is purpose-built for one job: generating the complete set of App Store Connect screenshots. It is not a general Xcode control surface, and it is not competing with the general-purpose Xcode MCP servers, it complements them.

ToolScopeBest forApp Store screenshot output
storescreensNarrow. App Store screenshot capture, device-size routing, locale and appearance matrix, HTML preview gallery.Producing the final screenshot set for App Store Connect in one command.Yes. Named, organized by device and locale, ready to upload.
[Apple Xcode MCP](https://developer.apple.com/documentation/xcode/giving-agentic-coding-tools-access-to-xcode) (built into Xcode 26.3+)Xcode-resident tools. Most notably RenderPreview for a single SwiftUI #Preview.Checking one view's layout without spinning up a simulator.No.
[XcodeBuildMCP](https://github.com/getsentry/XcodeBuildMCP)General iOS/macOS build, test, and device interaction driven by xcodebuild.Letting an agent compile, test, and debug iOS/macOS projects through a unified MCP interface.No.
[xc-mcp](https://github.com/conorluddy/xc-mcp)29 tools covering build, simulator lifecycle, and accessibility-first UI automation. Optimized for low-context agent interactions.Agents that need to drive the simulator via semantic accessibility queries (fast, token-cheap) instead of parsing screenshots.No, its screenshot tool is for a single capture, not a full App Store matrix.

If you are shipping an app, you will likely use StoreScreens alongside one of the general servers: the general server handles build and run, StoreScreens handles the screenshot matrix at the end.

Each run produces a browsable HTML preview with per-device galleries:

Preview index page listing every captured device

iPad Pro 13" gallery with all 19 screenshots

iPhone 6.9" gallery with all 19 screenshots

When the MCP server is configured, the agent streams per-screenshot progress inline as each device captures:

Capture starting - MCP tool called, taskId returned, polling begins

Per-screenshot progress streaming inline as each device captures

<video src="https://github.com/user-attachments/assets/fb0c0cf1-8fdc-4e28-98c9-1baded6dd947" controls></video>

🎯 aiskill88 AI 点评 A 级 2026-05-23

该工具提供了开源的MCP功能,适用于开发者捕获iOS和macOS应用截图,但功能和使用场景需要进一步完善

⚡ 核心功能
👥 适合人群
Claude Desktop / Claude Code 用户AI 工具开发者需要扩展 AI 能力的专业人士自动化工程师
🎯 使用场景
  • 在 Claude Desktop 对话中直接调用本地工具,实现 AI 与系统的深度联动
  • 通过自然语言驱动复杂的多步骤自动化任务,代替繁琐手动操作
  • 将多个 MCP 工具组合使用,构建个人专属 AI 工作站
⚖️ 优点与不足
✅ 优点
  • +MIT 协议,可免费商用
  • +标准化 MCP 协议,生态互联性强
  • +与 Claude 官方生态无缝对接
  • +即插即用,配置简单快捷
⚠️ 不足
  • 依赖 Claude 客户端,非 Claude 用户无法使用
  • MCP 协议仍在持续演进,接口可能变更
  • 需要一定的配置步骤
⚠️ 使用须知

AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。

📄 License 说明

✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。

🔗 相关工具推荐
🧩 你可能还需要
基于当前 Skill 的能力图谱,自动补全的工具组合
❓ 常见问题 FAQ
解答
💡 AI Skill Hub 点评

AI Skill Hub 点评:开源MCP工具:捕获iOS和macOS应用截图 的核心功能完整,质量良好。对于Claude Desktop / Claude Code 用户来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。

⬇️ 获取与下载
⬇ 下载源码 ZIP

✅ MIT 协议 · 可免费商用 · 直接从 aiskill88 服务器下载,无需跳转 GitHub

📚 深入学习 开源MCP工具:捕获iOS和macOS应用截图
查看分步骤安装教程和完整使用指南,快速上手这款工具
🌐 原始信息
原始名称 storescreens-cli
Topics app-storecliiosmacosswift
GitHub https://github.com/ciscoriordan/storescreens-cli
License MIT
语言 Swift
🔗 原始来源
🐙 GitHub 仓库  https://github.com/ciscoriordan/storescreens-cli 🌐 官方网站  https://storescreens.app

收录时间:2026-05-23 · 更新时间:2026-05-23 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。