Asc Localize Metadata logo

Asc Localize Metadata

Community
hanamizuki
asc-localize-metadata

Automatically translate and sync App Store metadata (description, keywords, what's new, subtitle) to multiple languages using LLM translation and asc CLI. Use when asked to localize an app's App Store listing, translate app descriptions, or add new languages to App Store Connect.

Overview

Publisherhanamizuki
Repositorysolopreneur
Skill nameasc-localize-metadata
Stars
150
Forks
9
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by hanamizuki on GitHub. Read the source before you install it.

Installation

Install the Asc Localize Metadata AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/hanamizuki/solopreneur.git /tmp/solopreneur
mkdir -p .claude/skills
cp -r /tmp/solopreneur/plugins/claude/ios-dev/skills/asc-localize-metadata .claude/skills/asc-localize-metadata
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Asc Localize Metadata in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Asc Localize Metadata on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Asc Localize Metadata is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

asc localize metadata

Use this skill to pull English (or any source locale) App Store metadata, translate it with LLM, and push translations back to App Store Connect — all automated.

Command discovery and output conventions

  • Always confirm flags with --help for the exact asc version:
    • asc localizations --help
    • asc localizations download --help
    • asc localizations upload --help
    • asc apps info edit --help
  • Prefer explicit long flags (--app, --version, --version-id, --type, --app-info).
  • Output defaults are TTY-aware: table in interactive terminals and JSON in CI or other non-interactive contexts. Use explicit --output when the format matters.
  • Prefer deterministic ID-based operations. Do not "pick the first row" via head -1 unless the user explicitly agrees.

Preconditions

  • Auth configured (asc auth login or ASC_* env vars)
  • Know your app ID (asc apps list to find it)
  • At least one locale (typically en-US) already has metadata in App Store Connect

Supported Locales

App Store Connect locales for version and app-info localizations:

ar-SA, bn-BD, ca, cs, da, de-DE, el, en-AU, en-CA, en-GB,
en-US, es-ES, es-MX, fi, fr-CA, fr-FR, gu-IN, he, hi, hr,
hu, id, it, ja, kn-IN, ko, ml-IN, mr-IN, ms, nl-NL, no,
or-IN, pa-IN, pl, pt-BR, pt-PT, ro, ru, sk, sl-SI, sv,
ta-IN, te-IN, th, tr, uk, ur-PK, vi, zh-Hans, zh-Hant

Two Types of Metadata

Version Localizations (per-release)

Fields: description, keywords, whatsNew, supportUrl, marketingUrl, promotionalText

App Info Localizations (app-level, persistent)

Fields: name, subtitle, privacyPolicyUrl, privacyChoicesUrl, privacyPolicyText

Workflow

Step 1: Resolve IDs

bash
# Find app ID
asc apps list --output table

# Find latest version ID
asc versions list --app "APP_ID" --state READY_FOR_DISTRIBUTION --output table
# or for editable version:
asc versions list --app "APP_ID" --state PREPARE_FOR_SUBMISSION --output table

# Find app info ID (for app-level fields like name/subtitle)
asc apps info list --app "APP_ID" --output table

Notes:

  • Version-localization fields (description, keywords, whatsNew, etc.) are per-version.
  • App-info fields (name, subtitle, privacy URLs/text) are app-level and use --type app-info.
  • If you only have names (app name, version string) and need IDs deterministically, use asc-id-resolver.

Step 2: Download source locale

bash
# Download version localizations to local .strings files
# (description, keywords, whatsNew, promotionalText, supportUrl, marketingUrl, ...)
asc localizations download --version "VERSION_ID" --path "./localizations"

# Download app-info localizations to local .strings files
# (name, subtitle, privacyPolicyUrl, privacyChoicesUrl, privacyPolicyText, ...)
asc localizations download --app "APP_ID" --type app-info --app-info "APP_INFO_ID" --path "./app-info-localizations"

This creates files like ./localizations/en-US.strings and ./app-info-localizations/en-US.strings. If download is unavailable, read fields individually:

bash
# List version localizations to see existing locales and their content
asc localizations list --version "VERSION_ID" --output table

Step 3: Translate with LLM

For each target locale, translate the source text. Follow these rules:

Translation Guidelines
  • Tone & Register: Always use formal, polite language. Use formal "you" forms where the language distinguishes them (Russian: «вы», German: «Sie», French: «vous», Spanish: «usted», Dutch: «u», Italian: «Lei», Portuguese: «você» formal, etc.). App Store descriptions are professional marketing copy — never use casual or informal register.
  • description: Translate naturally, adapt tone to local market. Keep formatting (line breaks, bullet points, emoji). Stay within 4000 chars.
  • keywords: Do NOT literally translate. Research what users in that locale would search for. Comma-separated, max 100 chars total. No duplicates, no app name (Apple adds it automatically).
  • whatsNew: Translate release notes. Keep it concise. Max 4000 chars.
  • promotionalText: Translate marketing hook. Max 170 chars. This can be updated without a new version.
  • subtitle: Translate or adapt tagline. Max 30 chars — this is very tight, may need creative adaptation.
  • name: Usually keep the original app name. Only translate if the user explicitly asks. Max 30 chars.
LLM Translation Prompt Template

For each target locale, use this approach:

Translate the following App Store metadata from {source_locale} to {target_locale}.

Rules:
- description: Natural, fluent translation. Preserve formatting (line breaks, bullets, emoji). Max 4000 chars.
- keywords: Do NOT literally translate. Choose keywords native speakers would search for in the App Store. Comma-separated, max 100 chars total. Do not include the app name.
- whatsNew: Translate release notes naturally. Max 4000 chars.
- promotionalText: Translate marketing tagline. Max 170 chars.
- subtitle: Adapt tagline creatively to fit 30 chars max.
- name: Keep the original app name unless explicitly requested to translate it. Max 30 chars.
- Use formal, polite language and formal "you" forms (Russian: вы, German: Sie, French: vous, Spanish: usted, Dutch: u, etc.). App Store copy is professional marketing — never use informal register.
- Respect cultural context. A playful tone in English may need adjustment for formal markets (e.g., ja, de-DE).

Source ({source_locale}):
description: """
{description}
"""

keywords: {keywords}

whatsNew: """
{whatsNew}
"""

promotionalText: {promotionalText}

name: {name}

subtitle: {subtitle}

Step 4: Upload translations

Option A: Via .strings files (bulk)

Create a .strings file per locale in the appropriate directory.

Version localization example:

// nl-NL.strings
"description" = "Je app-beschrijving hier";
"keywords" = "wiskunde,kinderen,tafels,leren";
"whatsNew" = "Bugfixes en verbeteringen";
"promotionalText" = "Leer de tafels van vermenigvuldiging!";

Then upload version localizations:

bash
asc localizations upload --version "VERSION_ID" --path "./localizations"

App-info localization example:

// nl-NL.strings
"subtitle" = "Leer tafels spelenderwijs";

Then upload app-info localizations:

bash
asc localizations upload --app "APP_ID" --type app-info --app-info "APP_INFO_ID" --path "./app-info-localizations"
Option B: Via individual commands (fine control)
bash
# Version localization fields (fine control).
# Prefer passing the explicit version ID for determinism.
asc apps info edit --app "APP_ID" --version-id "VERSION_ID" --locale "nl-NL" \
  --description "Je beschrijving..." \
  --keywords "wiskunde,kinderen,tafels" \
  --whats-new "Bugfixes en verbeteringen"

For app-level fields:

bash
# Subtitle/name (app-info localization) is managed via app-info localizations.
# Use the app-info localization .strings + upload flow; there is no app-infos localizations command.
#
# 1) Edit: ./app-info-localizations/nl-NL.strings
# "subtitle" = "Leer tafels spelenderwijs";
#
# 2) Upload:
asc localizations upload --app "APP_ID" --type app-info --app-info "APP_INFO_ID" --path "./app-info-localizations"

Step 5: Verify

bash
# Check all locales are present
asc localizations list --version "VERSION_ID" --output table

# Check app info localizations
asc localizations list --app "APP_ID" --type app-info --app-info "APP_INFO_ID" --output table

Character Limits (enforce before upload!)

FieldLimit
Name30
Subtitle30
Keywords100 (comma-separated)
Description4000
What's New4000
Promotional Text170

Always validate translated text fits within limits before uploading. Truncated text looks unprofessional. If translation exceeds the limit, shorten it — do not truncate mid-sentence.

Full Example: Add nl-NL and ru to Roxy Math

bash
# 1) Resolve IDs deterministically (do not auto-pick the "first" row)
# If you only have names, use asc-id-resolver skill.
asc apps list --output table
APP_ID="APP_ID_HERE"

asc versions list --app "$APP_ID" --state PREPARE_FOR_SUBMISSION --output table
VERSION_ID="VERSION_ID_HERE"

asc apps info list --app "$APP_ID" --output table
APP_INFO_ID="APP_INFO_ID_HERE"

# 2) Download English source (or your chosen source locale)
asc localizations download --version "$VERSION_ID" --path "./localizations"
asc localizations download --app "$APP_ID" --type app-info --app-info "$APP_INFO_ID" --path "./app-info-localizations"

# 3) Read en-US.strings, translate to nl-NL and ru (LLM step)

# 4) Write nl-NL.strings and ru.strings to:
#    - ./localizations/ (version localization fields)
#    - ./app-info-localizations/ (subtitle/name/privacy fields)

# 5) Upload all
asc localizations upload --version "$VERSION_ID" --path "./localizations"
asc localizations upload --app "$APP_ID" --type app-info --app-info "$APP_INFO_ID" --path "./app-info-localizations"

# 6) Verify
asc localizations list --version "$VERSION_ID" --output table
asc localizations list --app "$APP_ID" --type app-info --app-info "$APP_INFO_ID" --output table

Agent Behavior

  1. Always start by reading the source locale — never translate from memory or assumptions.
  2. Check existing localizations first — don't overwrite existing translations unless the user asks to update them.
  3. Version vs app-info is different — version fields live under --version "VERSION_ID"; subtitle/name/privacy live under --app ... --type app-info.
  4. Prefer deterministic IDs — do not select IDs via head -1 unless explicitly requested; use --output table for selection or asc-id-resolver.
  5. Validate character limits before uploading. Count characters for each field. If over limit, re-translate shorter.
  6. Keywords are special — do not literally translate. Research locale-appropriate search terms. Think like a user searching the App Store in that language.
  7. Show the user translations before uploading — present a summary table of all fields × locales for approval. Do not push without confirmation.
  8. Process one locale at a time if translating many languages — easier to review and catch errors.
  9. If upload fails for a locale, log the error, continue with other locales, report all failures at the end.
  10. For updates to existing localizations — download current, show diff of what will change, get approval, then upload.

Notes

  • Version localizations are tied to a specific version. Create the version first if it doesn't exist.
  • promotionalText can be updated anytime without a new version submission.
  • whatsNew is only relevant for updates, not the first version.
  • Use asc-id-resolver skill if you only have app/version names instead of IDs.
  • Use asc-metadata-sync skill for non-translation metadata operations.
  • For subscription/IAP display name localization, use asc-subscription-localization skill instead.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Asc Localize Metadata AI skill do?

Automatically translate and sync App Store metadata (description, keywords, what's new, subtitle) to multiple languages using LLM translation and asc CLI. Use when asked to localize an app's App Store listing, translate app descriptions, or add new languages to App Store Connect.

Why use Asc Localize Metadata on TypingMind?

Because you install it once and use it with any model. Asc Localize Metadata is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Asc Localize Metadata in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/hanamizuki/solopreneur/tree/main/plugins/claude/ios-dev/skills/asc-localize-metadata. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Asc Localize Metadata?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Asc Localize Metadata?

As many as you like. As long as a model supports skills, you can use Asc Localize Metadata with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Asc Localize Metadata AI skill free?

Yes. It is published on GitHub by hanamizuki under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇