- Home
- Skills
- Mobile Development
- App Store Listing Writer
Works with the AI Agents you already use
App Store Listing Writer
Write an App Store name, subtitle and keyword field that rank, with Apple's limits and wasted keywords checked by a script. Offline, no API keys.
Free
App Store Listing Writer
Example session with this skill installed
Write an App Store listing for my pomodoro timer app called FocusFlow. It has custom sounds, iCloud sync and a task manager. Competitors are Forest and Be Focused. Here is my current draft: name "FocusFlow: Pomodoro Timer", subtitle "Pomodoro focus timer & tasks", keywords "pomodoro, focus,timer,app,study,task,tasks,productivity".
- Read your context and instructions
- Compiled the app store listing
FocusFlow's draft wastes almost half its keyword field. I checked it first, then wrote three options targeting pomodoro timer, focus timer, study timer, task planner and white noise. Forest and Be Focused own the broad terms, so the options lean on specific phrases a new app can reach.
Your current draft
$ python3 scripts/check_listing.py draft.json
name 25 / 30 5 left
subtitle 28 / 30 2 left
keywords 55 / 100 45 left
PROBLEMS (must fix)
- keywords: spaces around commas waste characters (1x); write 'a,b,c'.
TIPS
- keywords: 'app' is wasted: every result is an app.
- 'pomodoro' is in name and subtitle and keywords; Apple combines these fields, so once is enough. Remove it from keywords to save 9 characters.
- 'timer' is in name and subtitle and keywords; Apple combines these fields, so once is enough. Remove it from keywords to save 6 characters.
- 'focus' is in subtitle and keywords; Apple combines these fields, so once is enough. Remove it from keywords to save 6 characters.
- 'tasks' is in subtitle and keywords; Apple combines these fields, so once is enough. Remove it from keywords to save 6 characters.
- both 'task' and 'tasks': one form is usually enough.
- target terms not covered by name, subtitle or keywords: white noise
Option 1 (recommended): pomodoro first
- Name: FocusFlow: Pomodoro Timer
- Subtitle: Study focus & task planner
- Keywords: work,productivity,tracker,deep,session,interval,break,todo,list,adhd,white,noise,sounds,sync,ambient
Idea: the main keyword in the name [meta-name-brand]; study, focus and task planner in the subtitle; the keyword field holds only new words [kw-no-repeat].
$ python3 scripts/check_listing.py option1.json
name 25 / 30 5 left
subtitle 26 / 30 4 left
keywords 100 / 100 0 left
No problems found.
Option 2: focus timer first
- Name: FocusFlow: Focus Timer
- Subtitle: Pomodoro, task & study sounds
- Keywords: productivity,work,deep,session,interval,break,todo,list,planner,adhd,white,noise,ambient,tracker
$ python3 scripts/check_listing.py option2.json
name 22 / 30 8 left
subtitle 29 / 30 1 left
keywords 96 / 100 4 left
No problems found.
Option 3: study timer first
- Name: FocusFlow - Study Timer
- Subtitle: Pomodoro focus with tasks
- Keywords: work,productivity,deep,session,interval,break,todo,list,planner,task,adhd,white,noise,ambient,sync
$ python3 scripts/check_listing.py option3.json
name 23 / 30 7 left
subtitle 25 / 30 5 left
keywords 98 / 100 2 left
TIPS
- both 'task' and 'tasks': one form is usually enough.
The tip is kept on purpose: "task" in the keyword field is what covers "task planner" in this option.
Recommendation
Option 1. "Pomodoro timer" is the term people use for this kind of app, and the name carries the most weight [kw-title-weight]. It uses all 100 keyword characters with no repeats.
Next steps
- Name, subtitle and keywords change only with a new app version [meta-version-bound]; paste them when you prepare the next one.
- Keep the words of any term you already rank top 10 for [kw-top10].
- Judge the change after one to two weeks [kw-measure].
Connects securely to your tools. The creator never sees your data.
About this skill
Writing an App Store listing is a puzzle: 30 characters for the name, 30 for the subtitle, 100 for the hidden keyword field, and every repeated word or space after a comma wastes characters Apple would have indexed.
What it does
- Learns the app first. Reads what the app really does from you or the repo (README, Info.plist, pubspec.yaml) and never claims features that do not ship.
- Picks words that can win. Keeps only relevant terms, prefers specific phrases a new app can rank for, and places them by weight: name, then subtitle, then keyword field.
- Writes three options for name, subtitle and keyword field, each with the idea behind it, plus a description, promotional text and What's New.
- Checks every option with a script. Character limits, spaces around commas, words repeated across fields, singular and plural of the same word, "app" in the keyword field, and target terms you forgot to cover.
- Cites its rules. Every choice points to one of 32 short ASO rules bundled with the skill, each with its source.
How it works
- Asks about the app, its competitors, its markets and any terms it already ranks for.
- Builds a list of 15–30 candidate terms and keeps the relevant, winnable ones.
- Writes three options and saves each as JSON.
- Runs
scripts/check_listing.pyon each option, fixes every problem and acts on the tips. - Recommends one option and gives copy-ready text per field.
Why this beats prompting it yourself
Apple indexes the name, subtitle and keyword field together, so a word used once anywhere is enough. A plain prompt repeats words across fields, adds "app", and miscounts characters. Here a script does the counting and catches the waste before you paste anything.
Use cases
- Writing the first listing for a new iOS app.
- Reworking an existing listing that does not rank.
- Writing What's New and promotional text for an update.
- Preparing a listing for a new language, researched per language instead of translated.
Known limitations
No live search-volume data and no connection to App Store Connect: you paste the result yourself. Localized listings still need a native speaker's check. For daily rank tracking and pushing changes to App Store Connect after approval, the same rules power Rankwell, a free open-source Claude Code plugin: https://github.com/AkashShahriar55/rankwell
Requires Python 3 for the validation script. Cannot pull live search volume data. Does not push changes directly to App Store Connect.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 9 days ago
- Free with an Agensi account
Requires
Permissions
File Scopes
Reads the app's README and project files to learn what the app does, writes each listing option as a small JSON file, and runs the bundled Python checker on it. No network access.