About AppRankHero
AppRankHero is an MCP server for app developers and marketers researching App Store and Google Play rankings, keywords, and competitors. It provides live access to team data through hosted or local connections.
Key features
- 01View app rankings and rank changes
- 02Review historical keyword trends
- 03Research keywords by country
- 04Compare apps competing for keywords
- 05Track or untrack keywords
- 06Manage tracked apps and reports
Use cases
- 01Find keywords with popularity and difficulty data
- 02Identify keywords that fell out of the top 10
- 03Review competitor apps for a search term
- 04Track keyword positions across countries
MCP support
Exposes AppRankHero account data and actions to MCP clients through a hosted Streamable HTTP endpoint or a local stdio server. Requires a Pro or Studio license.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://apprankhero.com/mcpclaude mcp add --transport http com-apprankhero https://apprankhero.com/mcp{"mcpServers":{"com-apprankhero":{"type":"http","url":"https://apprankhero.com/mcp"}}}Open in Cursor{"servers":{"com-apprankhero":{"type":"http","url":"https://apprankhero.com/mcp"}}}Open in VS Code{"mcpServers":{"com-apprankhero":{"type":"http","url":"https://apprankhero.com/mcp"}}}{"mcpServers":{"com-apprankhero":{"serverUrl":"https://apprankhero.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEAppRankHero MCP server
Your AppRankHero keyword rankings for the App Store (iOS) and Google Play (Android), inside ChatGPT, Claude, Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI and any other MCP client.
Ask in plain language:
- "Which of my keywords dropped out of the top 10 this week?"
- "Find popular, low difficulty keywords for a habit tracker in Germany."
- "Who outranks me for 'focus timer' and what do they have in their title?"
- "Track 'app blocker' and 'screen time' for my app in the US and UK."
Answers come live from your AppRankHero data, refreshed every morning.
Requirements
An AppRankHero Pro or Studio plan. Your license key starts with ARH-
and is in the Mac app under Account. See
plans and
how to activate a key.
Hosted server (recommended)
Nothing to install. One URL for every client:
https://apprankhero.com/mcp
Streamable HTTP, two ways to sign in:
- OAuth. ChatGPT, Claude and Perplexity open an AppRankHero page where you paste your license key once.
- Bearer header. Developer tools send
Authorization: Bearer ARH-XXXX.
Claude (claude.ai and Claude Desktop)
Settings, Connectors, Add custom connector, paste the URL, Connect.
ChatGPT
Settings, Apps, turn on Developer mode, Create, paste the URL, Authentication OAuth, Connect.
Claude Code
claude mcp add --transport http apprankhero https://apprankhero.com/mcp \
--scope user --header "Authorization: Bearer ARH-XXXX"
Codex
export ARH_TOKEN=ARH-XXXX
codex mcp add apprankhero --url https://apprankhero.com/mcp --bearer-token-env-var ARH_TOKEN
Cursor, Windsurf, VS Code and other JSON configs
{
"mcpServers": {
"apprankhero": {
"url": "https://apprankhero.com/mcp",
"headers": { "Authorization": "Bearer ARH-XXXX" }
}
}
}
VS Code uses "servers" instead of "mcpServers" and "type": "http".
Step by step guides for every client:
apprankhero.com/help/connect-ai-tools-mcp.
Local stdio server
For clients that only run local servers. Needs Node.js 18 or later.
{
"mcpServers": {
"apprankhero": {
"command": "npx",
"args": ["-y", "github:DedalDev/apprankhero-mcp"],
"env": { "ARH_LICENSE_KEY": "ARH-XXXX" }
}
}
}
| Variable | Required | Meaning |
|---|---|---|
ARH_LICENSE_KEY | yes | Your Pro or Studio license key |
ARH_TEAM_ID | no | Team to act on when you belong to several (see arh_whoami) |
ARH_API_BASE | no | API base URL, default https://apprankhero.com/api/v1 |
The hosted and the local server offer the same tools.
Tools
| Tool | What it does | Changes data |
|---|---|---|
arh_whoami | Active team, role, plan, limits and all your teams | no |
arh_list_apps | Tracked apps from both stores; the id is the app_id for other tools | no |
arh_app_rankings | Current ranks for one app with 24h, 7d and 28d changes, popularity and difficulty | no |
arh_rank_history | Rank trend for an app and keyword (default 90 days, up to 730) | no |
arh_report | Daily, weekly or monthly report: gainers, losers, top 10 changes | no |
arh_search_keywords | Keyword research: popularity and difficulty (0 to 100) by country | no |
arh_competitors | Apps competing for a keyword in a country | no |
arh_list_tracked_keywords | Tracked keywords for an app with status (ranked, unranked, pending) | no |
arh_search_store_apps | Search the live App Store or Google Play to find an app's store id | no |
arh_track_keywords | Track up to 200 keywords for an app in a country | yes |
arh_untrack_keywords | Stop tracking keywords for an app in a country | yes |
arh_add_app | Add an App Store or Google Play app, or a research project | yes |
Every tool carries MCP annotations (readOnlyHint, destructiveHint,
idempotentHint, openWorldHint) so clients can ask before write calls.
Write tools need the owner or member role; viewers are read only.
Plan limits apply exactly as in the Mac app.
App Store and Google Play
platformisios(App Store, default) orandroid(Google Play) onarh_search_store_apps,arh_add_app,arh_search_keywordsandarh_competitors. Tracked apps already know their store.store_idis the numeric App Store id (6772967892) or the package name (com.example.app).- Ranks are tracked within the top 200 on the App Store and the top 30 on
Google Play. Rank
0means not ranked in that range.
Privacy and security
- The server only sees data of the teams your license key belongs to.
- Privacy Policy and Terms of Service.
Support
[email protected] and the Help Center.
Made by DedalDev. MIT licensed.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.