agentleFS
HumanleapHumanleap/agent-skills/skills/magicscreenshots/SKILL.md

magicscreenshots

Skill0 starsChanged 41 days agoMITView on GitHub

Search real App Store listing screenshots and preview videos for design inspiration, then generate, restyle or localize the user's own App Store screenshots with MagicScreenshots. Use for App Store screenshot design, listing research, localization and app preview references.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. Humanleap/magicscreenshots-mcp★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 1 is distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Reference real results: Do not invent apps, chart ranks, screenshot styles or history.

    line 25
  2. Very specific:

    Search references: Use list_directory_apps by query, category, chart (free or grossing) and style tag, setting platform: "ios" for App …

    line 37
  3. Very specific:

    Inspect the images: Read get_directory_app and get_directory_screens for a few candidates.

    line 38
  4. Very specific:

    Prepare the user's generation: After secure authentication and spending approval, use get_screenshots to identify the user's app, then …

    line 39
  5. Very specific:

    Restyle: use the chosen reference slug, for example {"source_app":{"query":"My App"},"inspiration_slug":"duolingo","limit":5}, after …

    line 57
  6. Very specific:

    Localization: use create_localize with the user's selected target_locales, such as ["JPN","DEU","FRA"], and poll for each locale's output.

    line 58
  7. Very specific:

    Preview reference: search with has_video: true.

    line 59

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MagicScreenshots

Research real App Store marketing images and preview videos from a library refreshed daily. Choose a reference, then generate or localize screenshots for the user's own app.

Installation

Install this skill with pnpm dlx skills add Humanleap/agent-skills --skill magicscreenshots. Connect remote MCP to https://www.magicscreenshots.com/api/mcp.

CLI alternative (Node 18+):

pnpm dlx magic-screenshots search --category finance --chart grossing
pnpm dlx magic-screenshots app duolingo
pnpm dlx magic-screenshots screens duolingo --out ./reference

Four Hard Rules — Read First

  1. Reference real results. Do not invent apps, chart ranks, screenshot styles or history.
  2. Use third-party assets as references. Their developers own them. Never ship another app's screenshots or preview footage as the user's work.
  3. Approve spending before starting. Confirm the paid operation and the human’s spending limit before create_restyle or create_localize; do not start chargeable work without approval. Show the overview afterward and call approve_restyle only after the human approves that design.
  4. Keep output truthful. Generated screenshots must represent the user's actual app; a pending job is unfinished.

Authentication

Public library search/read tools need no key. For authenticated features, the human sets up a supported connection or secure CLI configuration using the official agent setup page. Never request, print, save or transmit API keys, passwords or authentication codes in conversation or tool arguments. Do not create credentials on the human’s behalf. If the assistant platform cannot provide secure authentication, use public library reads and hand the requested paid workflow to the official website. The human completes sign-in and payment there.

Core Workflow

  1. Define the brief. Identify the app category, audience and desired feel. Read list_directory_categories or list_directory_tags when needed.
  2. Search references. Use list_directory_apps by query, category, chart (free or grossing) and style tag, setting platform: "ios" for App Store research. Select relevant apps from actual results.
  3. Inspect the images. Read get_directory_app and get_directory_screens for a few candidates. Compare headline, colour, device treatment and order of benefits; cite the apps you inspected.
  4. Prepare the user's generation. After secure authentication and spending approval, use get_screenshots to identify the user's app, then create_restyle with its source_app and chosen inspiration_slug.
  5. Review and finish. Poll get_job_group for overview_url, show it, and call approve_restyle only after approval. Poll again for final output URLs.

Essential Tools

ToolsUse
list_directory_categories, list_directory_tagsDiscover categories and listing looks.
list_directory_appsSearch app listings; returned slugs identify apps.
get_directory_app, get_directory_screensListing details and ordered screenshots.
get_app_videos, get_app_historyPreview references or past listings; Pro access applies to downloads/history.
get_screenshotsIdentify the user's source app/screenshots.
create_restyle, create_localizeStart requested screenshot generation or translation.
get_job_group, approve_restyle, list_resultsInspect jobs, approve the overview and retrieve output.

Common Patterns

  • Design inspiration: compare 3–6 relevant apps and explain patterns visible in their actual screenshot sequences.
  • Restyle: use the chosen reference slug, for example {"source_app":{"query":"My App"},"inspiration_slug":"duolingo","limit":5}, after confirming the user's app identity.
  • Localization: use create_localize with the user's selected target_locales, such as ["JPN","DEU","FRA"], and poll for each locale's output.
  • Preview reference: search with has_video: true; compare opening seconds, pacing, captions and closing message. Do not reuse footage. Pro mp4 links expire after an hour.

Supporting Resources

Common Gotchas

  • The library contains App Store listing marketing screenshots, rather than arbitrary in-app UI captures.
  • A library reference is not evidence that the user's app offers the same features.
  • Show the returned claim/payment URL when a tool asks for access; do not guess entitlement or pay automatically.
  • An overview is a review stage, not the final rendered screenshot set.

Quick Reference

Search → inspect real references → identify user's app → create → show overview → human approves → retrieve completed output.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.