The Model Context Protocol is an open standard that lets AI agents call external tools. GetAppNiche runs a hosted MCP server, so your assistant can query live App Store data itself instead of guessing from its training set.

Yes. Calls are authenticated per key or per signed-in connection and scoped to your workspace; balance and per-request history live in your dashboard. Rotate or revoke a key any time it leaks, and disconnect an assistant in Settings → Connected apps.

◢ Model Context Protocol

Live App Store & Google Play data, right inside your AI

GetAppNiche is a real app-store-data MCP server. Connect Claude, ChatGPT, Cursor or any MCP client and your agent queries live store data itself — across the Apple App Store and Google Play — discovery, revenue estimates, ASO keyword difficulty and reviews, in one config.

Create an API key Setup Guides

✓ App Store + Google Play data · any MCP-capable agent · Streamable HTTP

Quick setup

All 14 setups ↓

bash · Claude Code

claude mcp add --transport http getappniche https://api.getappniche.com/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"

One command in your terminal — no restart needed.

json · Cursor

{
  "mcpServers": {
    "getappniche": {
      "url": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global).

toml · Codex CLI

[mcp_servers.getappniche]
url = "https://api.getappniche.com/mcp"
bearer_token_env_var = "GETAPPNICHE_API_KEY"

Add to ~/.codex/config.toml, then export GETAPPNICHE_API_KEY.

Video · 2 minutes

See it answer in Claude Code

Real Claude Code calls to the live server: connect with one command, find a niche by asking, score keywords, check what a rival changed — then the same key in Cursor, ChatGPT and other clients.

Video: Can you use App Store and Google Play data in Claude, ChatGPT or Cursor? — App Store & Google Play Data in Claude, ChatGPT & Cursor (MCP)2:03 · GetAppNiche on YouTube · Chapters and transcript →

Setup

Connect in under a minute

Pick your tool, paste one snippet, done — every guide uses the same API key from your dashboard. In claude.ai you sign in instead, no key needed.

1
Create an API key. Open Settings → API Keys, name it, and copy the secret — it is shown once.
2
Add the server. Paste your client’s snippet below, replacing YOUR_API_KEY with the secret.
3
Ask your agent. Try “Which meditation apps grew fastest last month?” — it picks the right tool itself.

Native HTTP One command in your terminal — no restart needed.

bash

claude mcp add --transport http getappniche https://api.getappniche.com/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"

In-app setup · no config file GetAppNiche is in Claude's connector directory: open claude.ai/directory/connectors/getappniche, click Connect and sign in to GetAppNiche — no API key, no config file. The steps below add the same server by hand as a custom connector (works in the Claude desktop app too; the free plan allows one custom connector).

In Claude, open Connectors (under Customize or Settings) and choose Add custom connector.
Name it GetAppNiche and paste the URL below as the MCP server URL.
Under Authentication, choose Sign in now — Claude may preselect No sign-in, so switch it. Keep the recommended OAuth client option.
Click Add, then Connect. Sign in to GetAppNiche and choose Allow access.
Ask your question in a chat. Claude asks before each tool call unless you allow the tools in the connector's settings.

connector url

https://api.getappniche.com/mcp

Native HTTP Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global).

json

{
  "mcpServers": {
    "getappniche": {
      "url": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

In-app setup · no config file Set up in ChatGPT's settings as a custom connector — Developer mode is required, and its availability depends on your ChatGPT plan.

In ChatGPT's settings, open the apps / connectors section and turn on Developer mode (it sits under the advanced options).
Create a new connector and name it GetAppNiche.
Paste the URL below as the MCP server URL.
Choose API key authentication and paste your GetAppNiche key.
In a new chat, enable the GetAppNiche connector and ask your question.

connector url

https://api.getappniche.com/mcp

via stdio bridge Add to claude_desktop_config.json, then fully restart the app. Rather sign in than paste a key? Use the Claude.ai steps — they work in the desktop app too.

json

{
  "mcpServers": {
    "getappniche": {
      "command": "npx",
      "args": ["-y", "@getappniche/mcp"],
      "env": { "GETAPPNICHE_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Native HTTP Add to ~/.codeium/windsurf/mcp_config.json (field is serverUrl).

json

{
  "mcpServers": {
    "getappniche": {
      "serverUrl": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Native HTTP Add to .vscode/mcp.json (Copilot agent mode).

json

{
  "servers": {
    "getappniche": {
      "type": "http",
      "url": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Native HTTP Add to ~/.codex/config.toml, then export GETAPPNICHE_API_KEY.

toml

[mcp_servers.getappniche]
url = "https://api.getappniche.com/mcp"
bearer_token_env_var = "GETAPPNICHE_API_KEY"

Native HTTP Add to ~/.gemini/settings.json (field is httpUrl).

json

{
  "mcpServers": {
    "getappniche": {
      "httpUrl": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

via stdio bridge Add to settings.json under context_servers.

json

{
  "context_servers": {
    "getappniche": {
      "source": "custom",
      "command": "npx",
      "args": ["-y", "@getappniche/mcp"],
      "env": { "GETAPPNICHE_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Native HTTP Cline panel → MCP Servers → Configure (cline_mcp_settings.json).

json

{
  "mcpServers": {
    "getappniche": {
      "type": "streamableHttp",
      "url": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Native HTTP Add to ~/.continue/config.yaml (headers under requestOptions).

yaml

mcpServers:
  - name: getappniche
    type: streamable-http
    url: https://api.getappniche.com/mcp
    requestOptions:
      headers:
        Authorization: Bearer YOUR_API_KEY

Native HTTP Settings → AI Assistant → MCP → Add → As JSON (2025.2+).

json

{
  "mcpServers": {
    "getappniche": {
      "url": "https://api.getappniche.com/mcp",
      "type": "streamable-http",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Native HTTP Point it at the endpoint with a Bearer header — or bridge stdio-only clients with npx @getappniche/mcp.

json

{
  "mcpServers": {
    "getappniche": {
      "type": "streamable-http",
      "url": "https://api.getappniche.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

# stdio-only client? Use the official bridge:
# GETAPPNICHE_API_KEY=YOUR_API_KEY npx -y @getappniche/mcp

MCP tools

Eight tools, priced in credits

The same intelligence as the dashboard, handed to your agent. Your plan’s API credits cover both.

search_apps 1 credit / row

Search the App Store and Google Play across 25+ filters — category, language, price model, rating, reviews, download and revenue estimates (min and max), release and update recency, and review-growth direction — sorted by any column.

get_app_detail 1 credit

Full detail for one app — metadata, rating, revenue and download estimates. Takes the canonical id (apple:284882215), a bare Apple id, an Android package name, or a store URL pasted from your browser.

get_app_historicals 1 credit

Historical metric time-series for a single app — reviews, ratings and estimates, daily then weekly, up to 730 days.

get_app_timeline 1 credit

One app's listing change history on the App Store or Google Play, newest first and grouped by day — new versions, icon and screenshot redesigns (visually verified), price and business-model changes, renames, category moves and description rewrites.

get_keyword_difficulty 10 credits

Deep ASO analysis for one keyword — popularity, difficulty and opportunity, with the top-ranking apps.

batch_keyword_difficulty 10 cr / kw

Score up to 10 keywords in a single call, auto-sorted so the best opportunities surface first.

get_app_reviews 1 credit / row

Live App Store or Google Play reviews for any app, tracked or not, in any storefront — or your monitored apps with sentiment, topics and improvement areas attached.

get_supported_countries Free

List every valid store country code before you plan keyword and market research. No credits.

Agent skills

Expert workflows, on demand

Drop a skill into your agent and it runs a battle-tested App Store framework — the tools above do the fetching.

App discovery

Turn a vague market question into a shortlist: state the thesis as filters — revenue band, review ceiling, launch recency — and rank what comes back.

Keyword research

Turn a niche into a prioritised keyword list: score candidates, prefer high opportunity at moderate difficulty, and return title/subtitle picks.

Metadata optimization

Draft title, subtitle and keyword-field variants with exact character counts, each tied to a keyword you already scored.

Competitor teardown

Deep-dive named competitors: trajectory from time-series, monetization, positioning — and what their own reviewers complain about.

Growth scan

Find the risers with a growth-direction query, then verify momentum against the time-series instead of trusting a snapshot.

Revenue benchmark

Map a niche's revenue distribution by sampling bands, not just the top ten, and set a realistic target for your own app.

Review mining

Cluster reviews of any App Store or Google Play app into themes, weight them by version and recency, and convert them into a ranked product to-do list.

Research brief

Maintain the shared context document — app, audience, competitors, goals — that every other skill reads before it starts.

install · claude code

npx skills add getappniche-com/aso-skills

Eight open-source skills — app discovery, keyword research, metadata optimization, competitor teardowns, review mining and more: github.com/getappniche-com/aso-skills. For Cursor and other clients, copy skills/ from the repo into .cursor/skills/. The server's protocol layer and a zero-dependency stdio bridge are open source: github.com/getappniche-com/mcp.

system prompt · drop-in fallback

Use GetAppNiche MCP for App Store and Google Play discovery,
competitor research, ASO keyword scoring, review signals and market sizing.

- Express the question as search_apps filters (revenue/download bands,
  review ceiling, rating floor, release and update recency, growth
  direction) instead of fetching broadly and filtering by eye.
- Call get_app_detail before claiming an app's downloads, revenue,
  rating or growth. It accepts a pasted store URL, so use the user's link.
- Use get_app_historicals when asked how a metric moved over time.
- Use get_app_timeline when asked what an app changed in its listing —
  versions, icon, screenshots, price, name, category, description.
- Score one keyword with get_keyword_difficulty; use
  batch_keyword_difficulty for 2–10 keywords.
- Read any App Store or Google Play app's reviews with get_app_reviews
  + app_id (an iOS id, or a package like google:com.duolingo); omit
  app_id for monitored apps to get sentiment and topics attached.
- Keep credit usage visible, and never invent data the tools didn't return.

Try it

Just ask your agent

Which meditation apps grew fastest last month?

Score keywords for a habit-tracker app in the US.

Give me a competitor brief for apple:284882215.

Find App Store or Google Play apps making $10K–50K/mo with under 1,000 reviews.

Which Health & Fitness apps grew more than 20% in the last 30 days?

Show me apps launched in the last 90 days already making over $5K/mo.

Compare “daily planner”, “focus timer” and “habit tracker” for ASO.

What are users complaining about in the top 3 budgeting apps?

Endpoint

https://api.getappniche.com/mcp

Transport: Streamable HTTP · JSON-RPC 2.0
Auth: Authorization: Bearer getappniche_… API key, or OAuth 2.1 sign-in (claude.ai)
Methods: initialize · tools/list · tools/call
Rate limit: 60 requests / minute per key
Credits: reported per call as credits_charged

Compatibility

If it speaks MCP, it works

Claude CodeClaude.aiCursorChatGPTClaude DesktopWindsurfVS CodeCodex CLIGemini CLIZedClineContinueJetBrains AI

Don’t see your tool? Point any Streamable-HTTP client at the endpoint with a Bearer header, or bridge stdio-only clients with npx @getappniche/mcp. Browse every setup →

FAQ

Frequently asked questions

What is MCP?

How much does it cost?

MCP shares the same credits as the REST API — 5,000 per month on Pro. search_apps and get_app_reviews cost 1 credit per row returned (50 rows a call by default with an API key, 20 when an assistant connects by signing in; up to 100); app detail, historicals and timeline cost 1 credit per call; keyword scoring costs 10 per keyword; listing supported countries is free. Every tool result reports the credits it charged.

Which AI tools can connect?

Anything that speaks MCP: claude.ai, Claude Code, Claude Desktop, ChatGPT, Cursor, Windsurf, VS Code, Codex, Gemini CLI, Zed, Cline, Continue, JetBrains and more. In claude.ai, GetAppNiche is listed in the connector directory — connect it there and sign in, no API key. ChatGPT connects as a custom connector in Developer mode, with your API key and no config file. Clients without native header support connect through our @getappniche/mcp stdio bridge (npx, zero dependencies). See every client setup →

How do I get an API key?

What data can the tools reach?

Is my usage private?

Give your AI agent App Store superpowers

Create a key, paste one snippet, and let your assistant research the market for you.

Start free — create a key Setup guides