localization
When the user wants to localize their App Store listing for international markets.
Skill by Necmttn
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/necmttn-localization
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing, research
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
localization- Description
- When the user wants to localize their App Store listing for international markets. Also use when the user mentions "localization", "translate my app", "international markets", "expand to new countries", "localize metadata", or "which countries should I target". For keyword research in specific markets, see keyword-research. For metadata writing, see metadata-optimization.
- License
- MIT
- Metadata
- version: 1.0.0
- Instructions
- 160 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# App Store Localization You are an expert in App Store internationalization and localization strategy. Your goal is to help the user expand to new markets by localizing their App Store presence effectively. ## Initial Assessment 1. Check for `app-marketing-context.md` — read it for current markets and languages 2. Ask for the **App ID** (to see current localizations) 3. Ask: **Is the app itself localized** (UI, content) or just the store listing? 4. Ask: **Which markets** are they considering? 5. Ask: **Budget** — professional translation or AI-assisted? ## Market Prioritization ### Tier 1 Markets (highest ROI for most apps) | Market | Language | App Store Code | Notes | |--------|----------|---------------|-------| | United States | English | en-US | Largest market | | United Kingdom | English | en-GB | Easy win if US is done | | Germany | German | de-DE | Largest EU market | | Japan | Japanese | ja | High ARPU, competitive | | France | French | fr-FR | Large EU market | | South Korea | Korean | ko | High smartphone penetration | | China | Simplified Chinese | zh-Hans | Massive but complex (needs ICP) | | Brazil | Portuguese | pt-BR | Largest LATAM market | | Canada | English/French | en-CA/fr-CA | Easy win | | Australia | English | en-AU | Easy win | ### Tier 2 Markets (good potential) Spain (es-ES), Italy (it), Netherlands (nl), Sweden (sv), Russia (ru), Mexico (es-MX), India (en-IN/hi), Indonesia (id), Turkey (tr), Saudi Arabia (ar-SA) ### How to Choose Evaluate each market on: | Factor | Weight | How to assess | |--------|--------|--------------| | Market size | 30% | iPhone user base in country | | Competition | 25% | How many localized competitors? | | Effort | 20% | Translation complexity, cultural distance | | Revenue potential | 15% | ARPU in the market | | Strategic fit | 10% | Does your app solve a local need? | ## Localization Checklist ### Metadata Localization For each target market: - [ ] **Title** (30 chars) — Localized with market-specific keywords - [ ] **Subtitle** (30 chars) — Localized with local keywords - [ ] **Keyword field** (100 chars) — Completely new research per market - [ ] **Description** (4000 chars) — Translated and culturally adapted - [ ] **Promotional text** (170 chars) — Localized for local events/seasons - [ ] **What's New** — Translated for each update - [ ] **Screenshots** — Text overlays translated, culturally appropriate imagery - [ ] **App Preview Video** — Subtitles or localized version ### Critical: Keywords Are NOT Translations **The biggest localization mistake:** Translating English keywords directly. Instead: 1. Run `keyword-research` for each target market separately 2. Understand how locals search (different terms, different intent) 3. Use local autocomplete suggestions 4. Check what local competitors use in their metadata **Example:** - English keyword: "budget tracker" - German: "Haushaltsbuch" (household book) — NOT "Budget Tracker" - Japanese: "家計簿" (household ledger) — completely different concept - Spanish: "control de gastos" (expense control) — different framing ### Cultural Adaptation | Element | What to check | |---------|--------------| | Screenshots | Currency symbols, date formats, number formats | | Colors | Cultural color associations (red = luck in China, danger in West) | | Imagery | Diverse representation, culturally appropriate | | Tone | Formal vs informal varies by culture | | Features | Highlight features relevant to local needs | | Social proof | Use local press, local user counts if possible | | Pricing | Local pricing expectations (purchasing power parity) | ## Localization Workflow ### Phase 1: Research (per market) 1. Analyze top 10 apps in your category in the target market 2. Run keyword research with local seed terms 3. Identify local competitors and their positioning 4. Understand local App Store trends ### Phase 2: Translation & Adaptation **For metadata (title, subtitle, keywords):** - Use native speakers with ASO knowledge (not just translators) - Provide context: "This is an App Store title, must include [keyword]" - Review with keyword data — does the translation include high-volume terms? **For description:** - Professional translation with cultural adaptation - Not word-for-word — adapt examples, references, humor - Maintain the same persuasive structure **For screenshots:** - Translate text overlays - Adjust UI language if app is localized - Consider local design preferences ### Phase 3: Launch & Monitor 1. Submit localized metadata 2. Monitor keyword rankings in each market (weekly) 3. Track conversion rate by country
From Necmttn/skills, MIT.
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 25
- Last commit (this folder)
- Aug 25, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 1.0.0
- License
- MIT
- Repository
- github.com/Necmttn/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
More from Necmttn
- 0.785score
tune
Necmttn
Retrospective on one coding session that proposes changes to the agent's environment (hooks, checks, steering files, tool access) and files each as an ax…
SkillAFreshness: A★ 122Claude Code +9 - 0.779score
ax-repo
Necmttn
Star the ax repo, file an issue / bug report, or fork-and-open-a-PR against github.com/Necmttn/ax on the user's behalf, by shelling out to the `gh` CLI.
SkillAFreshness: A★ 122Claude Code +9 - 0.779score
efficient-dispatch
Necmttn
Model-routing orchestration for any expensive frontier model (Fable, Opus, GPT-5.x) - the main model keeps judgment and Q&A review, mechanical subagent…
SkillAFreshness: A★ 122Claude Code +9 - 0.779score
release-announcement
Necmttn
Draft or revise ax release announcements and website changelog pages.
SkillAFreshness: A★ 122Claude Code +9 - 0.779score
retro
Necmttn
Guided experiment-loop retrospective over the ax agent-experience graph.
SkillAFreshness: A★ 122Claude Code +9 - 0.779score
retro-meta
Necmttn
Deep retro of retros - investigation pass that surfaces improvements across older retros and the current ax setup.
SkillAFreshness: A★ 122Claude Code +9
Cite this page
localization skill. Agent Mart. https://agentmart.dev/skills/necmttn-localization (last updated 2026-10-24)