MCP App Store
by KeenEthics
ASO Atlas icon

ASO Atlas MCP Server for ChatGPT

By KAROL JOZEF LABUDA

Overview

ASO Atlas helps app developers research keywords and track search rankings on the App Store and Google Play. Compare App Store keyword popularity and difficulty, explore related suggestions and competitor opportunities, and manage tracked apps, keywords and saved research lists. Google Play keywords show estimated demand bands instead of numeric popularity. Review metadata coverage and save private listing drafts without publishing to either store. With a connected App Store Connect account, inspect your own apps' private downloads, revenue and performance trends.

How to connect

  1. ChatGPT: Open the ASO Atlas listing in the ChatGPT app directory and choose Connect. ChatGPT listing ASO Atlas exposes 25 tools here.

Tools

add_competitor

ChatGPT
Add a competitor app to one of the user's tracked apps, in that app's storefront and store: an App Store competitor by its App Store id (itunes_id), a Google Play competitor (for a Google Play app) by its package name (store_id). It fetches the competitor's public listing from the store; adding a competitor that is already tracked is a no-op. Competitors power the keyword-gap analysis in get_app. Takes app_id from list_tracked_apps.

add_keyword_to_list

ChatGPT
Add a keyword to a keyword list in a given market (defaults to the list's own market) and store (App Store by default, or Google Play with platform "android"). A list is multi-market and multi-store, so the same term can be added for different storefronts and stores; a term already on the list for that storefront and store is not added twice. A term not measured yet is queued for measurement in the background. Takes list_id from list_keyword_lists.

add_keywords

ChatGPT
Track one or more keywords for an app in a given storefront. The same term can be tracked in multiple markets. Newly added keywords get their rank checked in the background. Takes app_id from list_tracked_apps.

app_ranking_keywords

ChatGPT
Reverse keyword lookup: the keywords an app currently ranks in the top results for in a storefront, with its position, popularity (0-100), difficulty (0-100) and when each position was captured. Pass app_id for one of your tracked apps (country defaults to its storefront), itunes_id for ANY App Store app, or store_id (a package name) with platform "android" for ANY Google Play app. Google Play rows carry a demand band instead of popularity. Coverage comes from our index and grows over time, so a small result set means limited coverage, not proof the app ranks nowhere. A null popularity with popularity_source 'below_floor' means demand exists but sits under the measurable threshold - report it as "<10".

change_app_market

ChatGPT
⚠️ Move an app to a different storefront. This re-points the app and migrates its tracked keywords to the new market, dropping the old market's ranking history. Use add_keywords instead if you only want to track terms in an extra market. Takes app_id from list_tracked_apps.

create_keyword_list

ChatGPT
Create a keyword list (a saved research collection) for the current user, optionally seeded with terms in a default market and store (App Store by default, or Google Play with platform "android"). Every call creates a new list, so repeating it makes a duplicate. Seed terms not measured yet are queued for measurement in the background. Returns the new list id.

get_app

ChatGPT
Get one tracked app in detail. Returns the app's storefront and rating (with the public-data synced_at), every tracked keyword (term, market, current & best rank with rank_checked_at, popularity, difficulty, and whether it's in the App Store Connect listing), the competitors tracked for it, and the keyword gap (terms competitors rank for that this app doesn't track yet). For a Google Play app every keyword's popularity is null and demand carries an estimated band (Very low to Very high) - present the band, never a number; current null there means the app is not in Google Play's results (Google shows one short list). A null popularity with popularity_source 'below_floor' is a measurement - demand exists but sits under the measurable threshold, report it as "<10", not as missing data. A keyword with pending true has no metric computed yet (research is queued) - that's "not measured yet", not zero. Takes app_id from list_tracked_apps.

get_app_optimizer

ChatGPT
Analyze a tracked app's App Store metadata for optimization in one storefront: returns the saved draft for that market, the app's last-synced live App Store listing for it (the live block, read from App Store Connect by the desktop app), the list of markets the app is active in, and an analysis of how well the metadata covers that market's tracked keywords (character budgets, missing terms ranked by keyword popularity, duplication). The analysis scores the draft when one exists, otherwise the live listing. When live.stale is true the public App Store already shows a newer listing than the synced one (live.store_title / live.store_synced_at carry the current public title and its sync time) - flag that the subtitle/keyword analysis may lag until the desktop app resyncs. A null popularity with popularity_source 'below_floor' is a measurement - demand exists but sits under the measurable threshold, report it as "<10", not as missing data. For the user's own App Store apps history.past_terms lists the terms this market's keyword field carried in earlier versions and no longer does (when each ran, the version that removed it, and the user's own retest verdict and note); in_draft true marks the ones the analyzed keyword field brings back - point those out as already tested, they are not new ideas. history.status "off" means the keyword history is not switched on (see get_keyword_history). Takes app_id from list_tracked_apps; country defaults to the app's primary storefront.

get_app_performance

ChatGPT
Get an app's private App Store performance digest from the user's connected App Store Connect account: the acquisition funnel (impressions, product page views, downloads, with view/conversion rates, the search-vs-browse source split and the top referrer apps and websites that sent the traffic), monetization (proceeds = all money earned incl. subscription payments, paid in-app purchases, paying subscribers vs free trials, MRR, new paying subscribers, trial starts and trial-to-paid conversions), health (sessions, installations, deletions, crashes), top territories, rating, and recent metadata changes with their measured before/after impact. Every summed metric covers the last N days its own source reported (analytics lags sales by 1-2 days, see period.data_through) and comes with a percent delta vs the prior period of the same length - null when history doesn't fully cover that prior period. Numbers and trends only, draw your own conclusions. Metadata-change impacts are correlational, not causal. Use this to diagnose questions like 'why aren't more users downloading or paying?', then get_performance_series to drill into any metric's daily history. Only works for apps in the user's own App Store Connect account. For a Google Play app in the user's linked Play Console the digest is Google's instead (platform: android): the store listing funnel (visitors, acquisitions = unique users who installed from the listing, conversion rate, traffic sources and the real Google Play search terms that led to the listing), installs (new users), uninstalls, active devices, crashes, ANRs, rating, countries, and listing changes with their impact. Google reports no search impressions and no revenue here, posts each day a few days late (see period.data_through; listing days after data_through.store_listing count as zero visitors, posted or not), gives a delta only when history covers the whole prior period, and folds small countries, sources and terms into 'Other'. Takes app_id from list_tracked_apps.

get_dashboard

ChatGPT
Get the ASO Atlas dashboard for the current user: for each tracked app, how many keywords sit in the top 10 / top 50, the 7-day keyword-visibility trend, recent App Store impressions, and the biggest ranking moves since the last check. Paged: returns up to limit apps (default 50) plus total; pass next_offset back as offset to read the next page, until it is null.

get_keyword_history

ChatGPT
The keyword-field history of one of the user's own App Store apps, read from App Store Connect: which keyword field every past version shipped with, what each version added and removed, and every term ever tried. Use it to answer "have I already tested this keyword, and how did it go?" before suggesting keyword-field changes. One localization per call (locale, e.g. "en-US"; defaults to the app's main one; locales lists the rest). versions are newest first: added / removed are against the version before, readded are added terms that had been used and dropped earlier, pending true is a version not on the App Store yet. affected pairs a change with the tracked keywords containing the changed words, with their rank up to 14 days before and after the release (before_checked / after_checked false = not being checked then, a null position with checked true = checked and not ranked; settled false = the release is younger than 14 days) - a hint, never proof: rank also moves for other reasons. terms is every term ever in the field: live false = removed, stints over 1 = removed and brought back, retest ("retest" or "skip") and note are the user's own verdict, set with set_keyword_note. date_exact false means the date is when the version was created in App Store Connect, some days before its release. status "off" means no key may read the listing yet - tell the user to turn keyword history on at settings_url (it needs an App Store Connect key with the Marketing role, or the desktop app); "pending" means the first read is running. Not available for Google Play apps or apps the user does not own. Takes app_id from list_tracked_apps.

get_keyword_list

ChatGPT
Get one keyword list with its keywords. Each keyword has its market and, when available, popularity and difficulty. A null popularity with popularity_source 'below_floor' is a measurement - demand exists but sits under the measurable threshold, report it as "<10", not as missing data. A keyword with pending true has no metric computed yet (research is queued) - that's "not measured yet", not zero; call again in a moment. A list is multi-market and multi-store: each row is a (keyword, storefront, store) triple. Google Play rows (platform "android") carry an estimated demand band and a null popularity - present the band, never a number. Takes list_id from list_keyword_lists.

get_performance_series

ChatGPT
Get the daily history of one App Store performance metric for a tracked app, from the user's connected App Store Connect account - optionally split by a dimension (source_type, territory, device, app_version, or app_referrer/web_referrer for the specific app or website that referred the traffic; only some metric/dimension combinations exist, an invalid one returns the valid dimensions). Use after get_app_performance to pinpoint when something changed (e.g. 'when did US impressions drop?'). Without a dimension, a null value means Apple reported no data that day (a gap, not zero). With a dimension, days without data are omitted and values beyond the top 8 are summed into 'other'. Only works for apps in the user's own App Store Connect account. A Google Play app in the user's linked Play Console has its own metrics instead: installs (new users), uninstalls, active_devices, visitors (store listing visitors), acquisitions (unique users who installed from the listing), conversion_rate, crashes, anrs, rating; installs split by territory, visitors and acquisitions by territory, source_type or search_term (the Google Play search terms that led to the listing). Google posts each day a few days late, so a Play window is the last N days ending on the last reported day; a day with no store listing data reads as 0 visitors whether Google has not posted it yet or nobody visited. Google's own bucket of small values is the column 'other (Google)'. Takes app_id from list_tracked_apps.

keyword_suggestions

ChatGPT
Get related keyword suggestions (store autocomplete) for a seed term in a storefront, on the App Store or Google Play - useful for discovering keyword ideas to research or track.

list_keyword_lists

ChatGPT
List the current user's saved keyword lists (research collections). Each entry has the list id (pass it to get_keyword_list as list_id), its name, default market, and keyword count.

list_tracked_apps

ChatGPT
List the apps the current ASO Atlas user tracks, on the App Store and Google Play. Each entry has the tracked-app id (pass it to other tools as app_id), the app name, its store (platform: ios or android), its store id (store_id: the App Store id, or the Google Play package name; itunes_id is set for App Store apps only), the storefront it's tracked in, and how many keywords are tracked for it. Paged: returns up to limit apps (default 100) plus total; pass next_offset back as offset for the next page, or narrow with query (part of the app name).

refresh_app

ChatGPT
Re-check an app's keyword rankings, ratings and reviews now (queues background jobs). Rankings already refresh automatically every night, and keywords checked within the last hour are skipped, so only call this after changing an app or its keywords, not to poll. Takes app_id from list_tracked_apps.

remove_competitor

ChatGPT
⚠️ Remove a competitor from a tracked app. Takes competitor_id (the "id" on a competitor from get_app).

remove_keyword

ChatGPT
⚠️ Stop tracking a single keyword and delete its ranking history. Takes keyword_id, which is the tracked-keyword id from get_app (the "id" on a keyword row).

remove_keyword_from_list

ChatGPT
⚠️ Remove a keyword from a keyword list. By default removes it from every market and store on the list; pass a country and/or platform to remove only those rows. Identify the keyword by its term. Takes list_id from list_keyword_lists.

research_keywords

ChatGPT
Research keywords in a storefront on the App Store (default) or Google Play (platform "android"). App Store: for each term, its popularity (search volume, 0-100), difficulty (0-100) and result count; a null popularity with popularity_source 'below_floor' is a measurement - demand exists but sits under the measurable threshold, report it as "<10", not as missing data. Google Play: popularity is always null and demand carries an estimated band (Very low to Very high) - present the band, never a number. Terms not measured yet are returned under pending while their metrics are computed - call again in a moment to get them. Nothing in the user's account changes (no tracked keywords, lists or search history); its only side effects are background jobs: terms not measured yet are queued for measurement, and estimated popularity values are queued for a more precise lookup. To keep terms, save them with create_keyword_list or add_keyword_to_list. Accepts up to 50 terms per call.

save_metadata_draft

ChatGPT
Save a draft of an app's store metadata for one storefront in ASO Atlas. App Store apps: title, subtitle, keywords (the keyword field). Google Play apps: title, subtitle (Play's short description) and description (the full description). This does NOT publish to either store - it only updates the draft used by get_app_optimizer. Only the fields you provide are updated, and they overwrite the previously saved draft values (no history is kept); pass an explicit empty string to clear one; fields the app's store does not have are ignored. Takes app_id from list_tracked_apps; country defaults to the app's primary storefront.

set_keyword_note

ChatGPT
Record the user's verdict on a term of their keyword history: whether it is worth testing again (retest: "retest" or "skip") and a short note on what happened when it ran. The note shows in the keyword history and next to the keyword field in the metadata optimizer when the term is typed back in. It replaces the previous verdict and note for that term; omit both retest and note to clear them. The term must be one from get_keyword_history for that locale. Takes app_id from list_tracked_apps.

track_app

ChatGPT
Start tracking an app for the current user in a storefront: an App Store app by its App Store id (itunes_id), or a Google Play app by its package name (store_id, e.g. "com.spotify.music") with platform "android". It fetches the app's public listing from the store and, when the user's App Store Connect account is connected, links the app's analytics in a background job. Idempotent: re-tracking an app you already track is a no-op. Returns the new tracked-app id.

untrack_app

ChatGPT
⚠️ Stop tracking an app and permanently delete it, its tracked keywords, ranking history and competitors from the account. Cannot be undone. Takes app_id from list_tracked_apps.

Capabilities

Writes

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/aso-atlas"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/aso-atlas)