agentleFS
Sign inStart free
ozdemirberkayozdemirberkay/seedshot/skills/verify/SKILL.md

verify

Skill0 starsChanged 20 days agoMITView on GitHub

Check App Store and Google Play screenshots before upload. Runs store checks (pixel sizes per store slot, alpha channels, images per set, screens missing from any language or appearance), then reviews every image for untranslated text, mixed languages, truncation, spinners, dialogs, debug banners and dark-mode problems, and writes a one-page HTML contact sheet. Use this when the user wants existing screenshots checked, reviewed or validated against store rules, or asks whether a screenshot set is ready to upload.

188 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use instinct-export, instinct-import and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • TheKREATORSSkill

    instinct-export

    Export instincts from project/global scope to a file

    Also in 3 projects that use this

  • TheKREATORSSkill

    instinct-import

    Import instincts from file or URL into project/global scope

    Also in 3 projects that use this

  • AviadPSkill

    risk-assess

    Assess the risk and blast radius of a proposed code change before implementing it. Evaluates CI impact, framework-wide vs localized scope, rollback path, and identifies blindspots. Use when user says "what's the risk", "is this safe", "assess the impact", "risk assess", "blast radius", or before applying framework-wide or CI-affecting changes. Do NOT use for code quality review (use code-quality) or plan validation (use design-validator).

    Also in 3 projects that use this

  • diego-tobalinaSkill

    build-fix

    Fix build and compilation errors systematically. Run when build fails.

    Also in 6 projects that use this

  • AaravChadhaSkill

    secure

    Security review with a confidence gate - a finding exists only with a concrete exploit scenario and a high/medium/low confidence rating; everything else goes to a worth-hardening list, never inflated. Sweeps five surfaces: auth gates, secrets hygiene, injection and unsafe sinks, unsafe deserialization/crypto/transport, and LLM tool-use trust boundaries. Reports only, never fixes. Use when the user asks for a security review or to check vulnerabilities, secrets, or auth.

    Also in 3 projects that use this

  • ArvorCoSkill

    specify

    Create feature specification from natural language description. Use when starting a new feature or creating a spec. Triggers on: create spec, specify feature, new feature spec.

    Also in 10 projects that use this

  • +1 more from TheKREATORS

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 167 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run python3 ${CLAUDE_PLUGIN_ROOT}/scripts/verify.py <output>.

    line 33
  2. Very specific:

    Add --fix to strip alpha channels.

    line 39
  3. Distinctive:

    Completeness: no spinners, skeletons, placeholder images, empty states, half-finished animations, keyboards, dialogs, toasts or debug …

    line 46
  4. Distinctive:

    Layout: no truncation, overflow or clipped text, which is common in German, Turkish, Finnish and Russian.

    line 49
  5. Very specific:

    If the project has seedshot mode (in the code, or saved in .seedshot/seedshot.patch), re-capture anything that fails, following …

    line 52
  6. Very specific:

    Run the final check with --sheet.

    line 56
  7. Distinctive:

    Report: Keep it short:

    line 60

“Common” is measured: kept by at least 6 in 10 of the 167 edited versions of this file we could read. “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

Seedshot: verify

The plugin folder is ${CLAUDE_PLUGIN_ROOT}. The files you read from it below aren't expanded: where their paths say CLAUDE_PLUGIN_ROOT, use this folder. The shell doesn't set that variable.

Before you start

Find the screenshot folder: the path the user gave, else output in .seedshot/config.json, else ask.

The checks expect Seedshot's layout:

<output>/<ios|android>/<phone|tablet>/<locale>/<light|dark>/<NN>-<screen-id>.png

If the images are in another layout (fastlane's <locale>/ folders, a hand-made set), mirror them into this layout as copies in a scratch folder and check those. Never move, rename or modify the user's own files. Say which copy --fix changed, so the user knows which files to upload.

Automated checks

Run python3 ${CLAUDE_PLUGIN_ROOT}/scripts/verify.py <output>. It checks:

  • pixel sizes against each store slot
  • alpha channels (both stores reject them)
  • per-folder counts (Apple ≤ 10, Play ≤ 8, Play tablets ≥ 4 recommended)
  • that every locale/appearance folder has the same screens.

Add --fix to strip alpha channels. Fix every error before delivering. ${CLAUDE_PLUGIN_ROOT}/references/store-specs.md has the rules and their sources.

Visual review: look at every image

It is tedious, and it is the step that makes Seedshot worth using. For each image, check:

  • Language: everything is in the folder's language. No mixed languages, no raw keys like home.title, and dates/numbers/currency are formatted for that locale.
  • Appearance: matches the folder (dark really is dark), and no unreadable text or invisible icons.
  • Completeness: no spinners, skeletons, placeholder images, empty states, half-finished animations, keyboards, dialogs, toasts or debug banners.
  • Access: signed in, and nothing locked: no lock icons, "PRO" tags on features, upgrade banners, trial countdowns or ads. Only the paywall shot, or a free-tier set the user asked for, may show them.
  • Status bar: clean (9:41, full battery and signal).
  • Layout: no truncation, overflow or clipped text, which is common in German, Turkish, Finnish and Russian.
  • Consistency: the same story as the other locales and devices.

If the project has seedshot mode (in the code, or saved in .seedshot/seedshot.patch), re-capture anything that fails, following ${CLAUDE_PLUGIN_ROOT}/skills/capture/SKILL.md for just those cells, then re-run verify.py. If it doesn't, list each failing image and what's wrong with it.

Contact sheet

Run the final check with --sheet. It also writes <output>/index.html, a grid of every screen across locales and appearances, so the user can review the whole set at a glance. Open it for them (open on macOS).

When run on its own

Report. Keep it short:

  • the folder checked and the contact sheet
  • image counts per platform/device/locale/appearance, and the store slots they fill
  • what was fixed, and what still needs the user (re-captures, real bugs in the app)
  • anything the user must do in the store consoles (e.g. "Play: tablet listing needs at least 4 images, you have 5 ✓"). If the images show features that need a subscription or purchase, remind them that the App Store description must say so (App Review Guideline 2.3.2).

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.

Check your setup, free