More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    store screenshots

    1

    Your agent designs App Store and Play Store screenshot strips with the open-source design_ss toolkit and renders them as store-size PNGs.

    Free

    Security scanned

    store screenshots

    Example session with this skill installed

    I need to generate App Store screenshots for my new fitness tracker app. I've got the raw screenshots ready, what's the next step?

    • Read your context and instructions
    • Compiled the store screenshots
    • Generated the document

    I'll set up design_ss so your agent can design your App Store screenshots.

    This is the first run, so design_ss isn't installed yet. I'll clone design_ss
    v0.1.6 into this skill's folder and install its dependencies, including a
    headless Chromium, about 350 MB, once. Shall I go ahead?

    This skill is installed inside your project, so add this line to your
    project's .gitignore to keep the download out of your repository:
    **/store-screenshots/design_ss/

    Once it's set up, put your files here:
    /Users/you/FitTrack/.claude/skills/store-screenshots/design_ss/input/

    WhatWhereRequired
    App briefinput/app.md: # FitTrack heading + ## SummaryYes
    iPhone screenshotsinput/iphone/, named for what they show (e.g. workout-log.png); 5+ is comfortableYes, for iPhone
    iPad screenshotsinput/ipad/Only for an iPad set
    App iconinput/appicon.pngOptional, helps the palette

    Tell me when the files are in place and I'll start the design.

    workout-log.png

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Mobile developers spend hours framing devices, writing headlines and exporting store screenshots at exactly the right sizes, then do it again for every device. This skill lets your own agent do that work with design_ss, an open-source toolkit that designs screenshot strips as HTML and renders them in headless Chromium at the sizes App Store Connect and Play Console accept.

    The skill is a coordinator. It sets up design_ss (with your permission), tells you what to put in its input folder, hands the design work to design_ss's own instructions, and gives you the path to the finished PNGs. Each run composes a fresh concept (palette, typeface, device placement, rhythm) rather than filling in a template.

    What it does

    • Setup: clones design_ss (pinned version) and installs its dependencies and a headless Chromium, once, after asking you.
    • Input check: confirms app.md has your app's name and summary and that at least one target folder has screenshots, and gives you the exact folder path when something is missing.
    • Design and render: your agent follows design_ss's design rules, checks the strip and renders store-size PNGs for iPhone, iPad, Android phone, and 7" and 10" Android tablets.
    • Copywriting: drafts any panel headlines you didn't write, based on your summary and what your screenshots show, and saves them in app.md for you to review. It never claims features your app doesn't have.
    • Hand edits: a local visual editor for adjusting layout, type and colour.

    How it works

    1. Install the skill and ask your agent for store screenshots.
    2. Set up (first time only): the agent asks, then downloads design_ss and its dependencies (~350 MB).
    3. Add your input: put app.md (name + short summary) and your screenshots in the input folder the agent shows you, one folder per device.
    4. Design: the agent designs, checks and renders the strip, iterating a few rounds.
    5. Get the PNGs from the output folder it gives you and upload them to App Store Connect or Play Console.

    Frameworks & tools

    Node.js, npm, Git and Playwright (headless Chromium). Runs locally; design and rendering work offline once set up.

    Why this beats prompting it yourself

    A chat can't hand you PNGs at the exact pixel sizes the stores require, with your real screenshots inside device frames. design_ss gives your agent the renderer, device frames, a schema checker and measured render feedback, so the agent checks its own output before it gives it to you.

    Use cases

    • Turn raw device captures into a full iPhone set for App Store Connect.
    • Design one consistent set across iPhone, iPad, Android phone and tablets.
    • Re-run with a new direction ("warmer palette", "lead with the timeline").
    • Make a targeted change to an existing set ("make panel 2's device bigger").

    Known limitations

    • Needs an agent that can run shell commands on your computer.
    • Node.js 20+ (22 recommended), Git, and ~350 MB of disk for the first setup.
    • Developed and tested on macOS. Linux should work. Windows is untested.
    • One app at a time per install; a re-run replaces the previous output, so copy the PNGs you want to keep.
    • Updating or reinstalling the skill can remove the downloaded design_ss folder, including your input and output.
    • Design quality and run cost depend on the agent and model you use.

    Requires Node.js (v20 minimum) and Git. Needs ~350MB disk space for headless Chromium. Processes one app project at a time.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 22 days ago

    • Free with an Agensi account

    Requires

    GitHub
    Npmjs
    Playwright
    Microsoft
    Node.js
    macOS
    Git

    Permissions

    Terminal / Shell
    Browser
    Read Files
    Write Files
    Network Access

    Allowed Hosts

    github.com
    registry.npmjs.org
    cdn.playwright.dev
    playwright.download.prss.microsoft.com

    File Scopes

    design_ss/**
    design_ss/input/**
    design_ss/strips/**
    design_ss/composer/references/history.md
    ~/Library/Caches/ms-playwright/**
    ~/.cache/ms-playwright/**
    ~/.npm/**
    Listed22 days ago

    What's inside

    Frequently Asked Questions

    The skill and design_ss are free and open source (MIT). The design work is done by your own agent, so a run uses your agent’s normal usage or API cost, typically a few render-and-review rounds per set. Setup, checking, rendering and the editor run locally at no cost.

    Any agent that supports SKILL.md skills and can run shell commands on your computer, such as Claude Code, Codex, Cursor or Gemini CLI. You need Node.js 20+ (22 recommended), Git and about 350 MB of disk. It’s developed and tested on macOS. Linux should work, and Windows is untested.

    You provide an app.md with your app’s name and a short summary, plus your real screenshots in one folder per device: iphone, ipad, phone, tablet_7 or tablet_10. You get one PNG per panel at the exact size the store accepts, ready to upload to App Store Connect or Play Console, plus a combined preview of the whole set.

    No. Any headline you write in app.md is used word for word. Headlines you leave out are drafted from your summary and screenshots, then saved back into app.md for you to review. Every claim must be supported by your summary or visible in a screenshot. That matters because Apple requires screenshots to show the real app.

    Only the first-time setup uses the network, and only after you agree: it downloads design_ss, its npm packages and a headless Chromium. Designing, checking and rendering run offline on your machine. Your screenshots are processed locally, though your agent’s model sees what the agent reads, as with any agent task. To change a result, ask for a specific edit (“make panel 2’s device bigger”), re-run with a new direction (“warmer palette”), or adjust it by hand in design_ss’s local visual editor.

    New in Mobile Development