Free $10 AppTweak API creditswhen you subscribe to Creator, within 7 days of signing up
Soku AI

Ask your AI which keywords an app wins, and what its users hate.

App Store and Google Play keyword rankings, search volume, and review sentiment — answered inside ChatGPT or Claude. You need no AppTweak subscription and no API key of your own: Soku buys each call, tells your AI the price before it runs, and charges the credits the call actually reported.

Start asking

Set it up in

Claude CodeCodexCursorVS CodeChatGPTClaude (web & desktop)Your own agent
The ask
New chatsoku · connected

Which keywords does this competitor app win in the App Store, and what are their users complaining about?

apptweak_app_keywordsAppTweak · Soku-suppliedcharged $0.0438 tracked terms; top 3 positions on 6 of them
apptweak_keyword_appsAppTweak · Soku-suppliedcharged $0.03You rank 14th on their strongest term; they are 2nd
apptweak_app_reviewsAppTweak · Soku-suppliedcharged $0.071-star reviews cluster on one thing: the export flow
They own the category term — you are 14th where they are 2nd — but their reviews say the same word over and over: export. The opening is not the keyword, it is the listing copy: their own users are describing the gap your app already closes.One store front and one country per call · store search volume is modelled, the stores publish none of it

Where this came from

Built-in — Soku buys itquoted first

  • AppTweak$0.14
3 calls$0.14

Illustrative figures. Every source, mark and price rule comes from the live Connect catalogue.

About five minutes

Set it up once, in the app you already use.

There is nothing to install for AppTweak specifically. One connection reaches every source in the catalogue, and this is one of them.

Claude Code

One sign-in covers every project, headless runs (`claude -p`, CI, the Agent SDK) work without a browser, and it can act on your accounts once you approve each change — the MCP connection is read-only.

  1. 1

    Create your Soku workspace

    The connection is public and identical for everyone — your Soku sign-in is what points it at your data. Connect one account in that workspace too: that is what your AI reads. The first time your AI uses the connection, Soku asks you to approve the access and names the workspace it will read through — so the workspace has to exist by then.

    Create your workspace

    Already have one? Nothing to do here — sign in when the connection asks for it.

  2. 2

    Copy the Soku MCP connection

    https://api.soku.ai/mcp
  3. 3

    Add it to Claude Code

    claude mcp add --transport http soku https://api.soku.ai/mcp
    1. Run the command in any project. It writes the configuration and prints an `Added …` line.
    2. Authorize separately — adding a server does not open a browser.
    3. Run `claude mcp list` to confirm the server reads `✔ Connected`.
  4. 4

    Approve the access, then ask it something

    1. Inside Claude Code, run `/mcp` and follow the browser flow — or, from your shell, `claude mcp login soku`.
    2. The Soku consent page names the host, the workspace the grant will read through, and the scopes it is asking for.
    3. Approving creates the grant listed above on this page. Cancelling creates nothing.
    4. On a machine with no browser (SSH, headless Linux), `claude mcp login soku --no-browser` prints the URL and takes the pasted callback.

    Paste this first

    Using Soku, search for what capabilities I can reach.

    A list of capabilities with names, what each reads and what each costs. An empty list means the grant carries no scopes — re-authorize and tick them. An error naming authorization means the host never completed the browser step.

Checked 2026-08-25 against Claude Code documentation. Menu names change between releases — look for connectors, apps, tools or integrations.

Open the console to connect your accounts
5 things you can ask it

Ask in plain English. These are the tools behind the answer.

Each one is a tool your AI is offered by name once you connect. You never have to type the name — it is here so you can tell what actually ran.

  • Which keywords an app actually wins

    apptweak_app_keywords

    Tracked keywords for one app with its ranking position on each — yours, or the competitor you are trying to displace.

    • “Which keywords does this app rank for?”
    • “Show me the terms this competitor wins in the App Store.”

  • What a store keyword is worth

    apptweak_keyword_volume

    Search volume, difficulty and competition for one keyword inside an app store — which is a different market from web search, with different winners.

    • “How much traffic does the app-store keyword photo editor get?”
    • “Is this term worth targeting, or is it already locked up?”

  • Who you are up against on a term

    apptweak_keyword_apps

    The apps currently ranking for one keyword, in position order — the competitive set for a term rather than for a category.

    • “Which apps rank for photo editor?”
    • “Who sits above us on this keyword right now?”

  • What reviewers complain about

    apptweak_app_reviews

    Reviews for one app with rating, text, date and sentiment — the fastest read on a competitor’s unfixed problem, in their users’ own words.

    • “What do users complain about in this app?”
    • “Summarise the negative reviews for this competitor since June.”

  • How an app presents itself in the store

    apptweak_app_metadata

    Title, subtitle, description, icon, category, rating and publisher for one listing on either store.

    • “Look up this app on the App Store.”
    • “What category is this app in, and how is it positioned?”
Before you rely on it

When AppTweak is the right tool — and when it is not.

Reach for it when

  • You are doing ASO and need an app’s current keyword positions rather than a monthly export.
  • You want to know which terms a competitor app wins, and who else is ranking on them.
  • You are choosing which app-store keywords to target and need volume before anyone rewrites a listing.
  • You are mining a competitor’s reviews for the complaint your positioning can answer.
  • You want a competitor’s store listing as it is right now, not as their press kit describes it.

Use something else when

  • You want WEB search rankings or volume. App-store search is a separate index with its own keywords and its own winners. See the alternatives.
  • You want your own installs, retention or revenue. That is your analytics or attribution account, read with your own credential and free. See the alternatives.
  • You want short-video or social performance for the app. That is a different kind of source entirely. See the alternatives.
  • You need every review ever left. Reviews come back per store front and date range, so a complete history is many calls, each one charged.
  • You need an exact install count. Store intelligence models what stores do not publish, and a model is not a measurement.
What it costsNeeds a paid plan

Quoted to your AI before the call runs, then charged on the credits AppTweak itself reports for that exact call.

Metered from the `metadata.request.cost` AppTweak returns inside its own response, at $0.000796 per credit — so a keyword lookup and a full review pull are not the same price, and neither is a guess. A response that reports no usable cost falls back to a $0.001 floor.

No fixed price table, on purpose: the credit cost comes back with the data, so an upstream reprice needs no change here. A call that fails is never charged, and a read AppTweak itself prices at zero credits is charged zero rather than rounded up to the floor.

Free $10 AppTweak API credits. Credited when you subscribe to Creator within 7 days of signing up. That covers dozens of competitor lookups.

Limits, stated up front

  • Everything is scoped to one store front at a time — `ios` or `android`, and one country. Two stores means two calls, and so does two markets.
  • Reviews are filtered by date range, so a long history is a sequence of charged calls rather than one.
  • Keyword volume and difficulty are AppTweak’s model of store search, which the stores themselves do not publish. Treat the ordering as reliable and the absolute numbers as estimates.
  • Responses wrap their data under a `content` / `metadata` envelope, so an agent reading a raw field path from the vendor’s docs may need one level more than it expects.
  • Built-in integrations need an active paid plan, because these are the ones Soku buys from a provider.
Two ways to get it

Buy AppTweak directly, or reach it through Connect.

AppTweak directly

  • An AppTweak plan sized to a credit allowance you commit to monthly
  • A key of your own to store, rotate and hand to an agent
  • A monthly bill that does not move when your usage does

Through Soku Connect

  • Per call, quoted before it runs, on the same credit balance as everything else
  • No AppTweak seat and no key to hand to an agent
  • Reachable from any MCP client, in the chat you are already in

What Connect does not give you

You get the store-intelligence reads listed on this page, not the AppTweak product. There is no workspace to curate, no historical dashboard, no alerting and no team seat — and every question is a call, so asking the same one weekly is charged weekly. This answers questions on demand; it does not watch anything for you. If you need the monitoring, buy the tool.

Upstream documentation: https://docs.apptweak.com/api