Token导航 LogoToken导航

Agent Skill · 研究检索

app-store-screenshot-studio(应用商店截图工作室)

app-store-screenshot-studio 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

app-store-screenshot-studio 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证信息还不明确,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 72,323。

总安装 188周安装 8GitHub Stars 16下载 66核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:app-store-screenshot-studio(应用商店截图工作室)
来源仓库:https://github.com/gitethanwoo/app-store-screenshot-studio
仓库路径:skills/app-store-screenshot-studio
安装命令:
npx skills add https://github.com/gitethanwoo/app-store-screenshot-studio --skill app-store-screenshot-studio
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/gitethanwoo/app-store-screenshot-studio --skill app-store-screenshot-studio

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

提供分步式工作流,用于制作 App Store Connect 审核就绪的标准截图集。适合需要严格遵循 Apple 设计规范与合规要求的开发者使用。按顺序完成每个步骤,包括定义截图目标、撰写文案和生成素材。必须完整执行当前阶段并获取用户确认后再进入下一环节。安装前建议核实本地环境是否具备图像处理与自动化部署能力。

来源仓库内路径:skills/app-store-screenshot-studio

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/gitethanwoo/app-store-screenshot-studio --skil…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

app-store-screenshot-studio 怎么安装?

在终端执行 npx skills add https://github.com/gitethanwoo/app-store-screenshot-studio --skill app-stor…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

app-store-screenshot-studio 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

app-store-screenshot-studio 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

app-store-screenshot-studio 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

app-store-screenshot-studio 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 16 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

app-store-screenshot-studio 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
app-store-screenshot-studio

应用商店截图工作室 · 本页

18816接触敏感数据未明确未通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

App Store Screenshot Studio

A step-by-step workflow for producing App Store Connect-ready screenshot sets.

Important: This is a sequential workflow. Complete each step fully and get user confirmation before moving to the next. Do not skip ahead.


Step 1: Create the Screenshot Brief

Goal: Define what screenshots to capture and what marketing copy to use.

What to do

Create a brief with 5-10 frames. For each frame, specify:

FieldDescription
frame_ide.g., 01_hero, 02_search
screenWhich app screen/state to capture
headline2-5 words, benefit-led
subheadlineOptional, max 10 words
compliance_notesIAP disclosure needed? Privacy wording?

Rules of thumb:

  • One idea per frame
  • Strongest value prop first
  • Keep text big and scannable
  • Use fictional/demo data only
  • Never misrepresent features the app doesn't have

See REFERENCE_APPLE.md for compliance guardrails.

Checkpoint

STOP and ask the user:

"Here's the screenshot brief. Does this capture the right screens and messaging? Any changes before I start capturing?"

Do NOT proceed to Step 2 until the user approves the brief.


Step 2: Capture Raw Screenshots

Goal: Get real screenshots from the app running in a simulator.

Gate: Verify Xcode Build MCP

Before proceeding, check if Xcode Build MCP tools are available (e.g., xcodebuild_build, simulator_boot, simulator_screenshot).

If NOT available:

STOP. I need Xcode Build MCP to capture screenshots from the simulator.

Install it with:
  claude mcp add XcodeBuildMCP npx xcodebuildmcp@latest

Or add to your MCP config (~/.claude.json):
  {
    "mcpServers": {
      "XcodeBuildMCP": {
        "command": "npx",
        "args": ["-y", "xcodebuildmcp@latest"]
      }
    }
  }

Alternatively, you can provide raw screenshots manually and I'll skip to Step 3.

What to do

  1. Build the app
  2. Boot the correct simulator (see STEP_2_CAPTURE.md for device selection)
  3. Navigate to each screen in the brief
  4. Capture screenshots

Output convention:

./screenshots/raw/<locale>/<device_category>/<frame_id>.png

Example: ./screenshots/raw/en-US/iphone_6_9/01_hero.png

See STEP_2_CAPTURE.md for detailed capture guidance.

Checkpoint

STOP and show the user the captured screenshots:

"Here are the raw screenshots I captured. Do these look correct? Any screens need to be recaptured?"

Do NOT proceed to Step 3 until the user confirms the raw screenshots are good.


Step 3: Generate Draft Composites (1K)

Goal: Create low-resolution marketing composites for approval before investing in high-res.

Gate: Verify Nano Banana Pro

Check if the Nano Banana Pro script exists:

ls ~/.claude/skills/nano-banana-pro/scripts/generate_image.py

If NOT available:

STOP. I need Nano Banana Pro to generate marketing composites.

Install it with:
  git clone --depth 1 https://github.com/steipete/agent-scripts /tmp/agent-scripts && \
    cp -r /tmp/agent-scripts/skills/nano-banana-pro ~/.claude/skills/ && \
    rm -rf /tmp/agent-scripts

Gate: Verify GEMINI_API_KEY

Check if the API key is set:

echo $GEMINI_API_KEY | head -c 10

If NOT available:

STOP. Nano Banana Pro requires a Gemini API key.

Set it with:
  export GEMINI_API_KEY="your-key-here"

Get a key at: https://aistudio.google.com/apikey

What to do

For each frame in the brief, generate a 1K draft:

uv run ~/.claude/skills/nano-banana-pro/scripts/generate_image.py \
  --input-image "./screenshots/raw/en-US/iphone_6_9/01_hero.png" \
  --prompt "Your compositing prompt here" \
  --filename "./screenshots/drafts/en-US/iphone_6_9/01_hero.png" \
  --resolution 1K

CRITICAL:

  • --input-image is MANDATORY. Never omit it.
  • Use --resolution 1K for drafts (faster, cheaper).
  • Output to ./screenshots/drafts/ (not final/).

See STEP_3_COMPOSITE.md for prompt templates.

Checkpoint

STOP and show the user the draft composites:

"Here are the 1K draft composites. Do these look good? Any changes to copy, layout, or style before I generate the high-res finals?"

Do NOT proceed to Step 4 until the user approves the drafts.


Step 4: Generate Final Composites (2K/4K)

Goal: Re-generate approved drafts at full resolution.

What to do

For each approved draft, regenerate at 2K (or 4K for iPad):

uv run ~/.claude/skills/nano-banana-pro/scripts/generate_image.py \
  --input-image "./screenshots/raw/en-US/iphone_6_9/01_hero.png" \
  --prompt "Same prompt as draft" \
  --filename "./screenshots/final/en-US/iphone_6_9/01_hero.png" \
  --resolution 2K

Output convention:

./screenshots/final/<locale>/<device_category>/<index>_<slug>.png

Validate before proceeding

Run these checks:

  • All images are accepted pixel sizes (see REFERENCE_SIZES.md)
  • Format is.png or.jpg
  • 1-10 images per device category
  • No prices or competitor references in copy
  • IAP disclosures present where needed

Checkpoint

STOP and show the user the final composites:

"Here are the final high-res composites. Ready to package for submission?"

Step 5: Package for Submission

Goal: Produce submission-ready assets.

What to do

  1. Create a ZIP of ./screenshots/final/
  2. Generate a manifest file:
{
  "locales": {
    "en-US": {
      "iphone_6_9": [
        "01_hero.png",
        "02_search.png",
        "03_detail.png"
      ]
    }
  }
}

Final summary

End your response with:

  1. What was generated (locales, device categories, counts)
  2. File tree showing raw + final paths
  3. Any compliance notes (IAP disclosures, etc.)
  4. Recommended next steps

Quick Reference

平台分布

Codex

33.31%
按下载量换算22

Claude

30.84%
按下载量换算20

Cursor

20.23%
按下载量换算13

Gemini CLI

9.71%
按下载量换算6

安全审计

Gen Agent Trust Hub

未通过

Socket

未通过

Snyk

可疑

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills