Getappniche Com MCP

What getappniche-com/mcp does

The getappniche-com/mcp MCP server exposes GetAppNiche’s App Store and Google Play intelligence to MCP-compatible agents. It is designed for research tasks that need current app listings, store metadata, estimated downloads or revenue, ASO keyword measurements, historical trends, and review data.

App discovery is handled by search_apps, which supports more than 25 filters. These include store, category, language, price model, rating, review count, downloads, revenue, release or update recency, and review growth. Results can be sorted by a selected column, with up to 100 apps returned per call. get_app_detail accepts a canonical app ID, an Apple ID, an Android package name, or a store URL.

How it works

The hosted MCP endpoint is available at https://api.getappniche.com/mcp and uses Streamable HTTP with JSON-RPC 2.0. Requests authenticate with a Bearer API key. Clients that cannot add HTTP headers can use the repository’s stdio bridge, available as a Node package through npx or as a dependency-free Python script. Both bridge options forward MCP messages to the hosted endpoint over TLS.

The protocol flow uses standard MCP initialization followed by tool listing and tool calls. The bridge supports the MCP session identifier and can handle JSON or SSE-framed responses. Tool-level failures, such as an invalid identifier or exhausted credits, are returned as tool results that the agent can inspect.

Setup and configuration

For direct HTTP connections, configure the client with the hosted URL and an Authorization: Bearer header. The README provides configuration examples for Claude Code, Cursor, Claude Desktop, VS Code, Windsurf, Codex CLI, Gemini CLI, Zed, Cline, Continue, and JetBrains AI. ChatGPT can connect through a custom connector in Developer mode using API-key authentication.

The local bridge reads the GETAPPNICHE_API_KEY environment variable. The Node option requires Node 18 or newer and can be started with npx -y @getappniche/mcp; the Python option requires Python 3.9 or newer and uses only the standard library. The API key is sent in the authorization header to the GetAppNiche endpoint.

Tools and capabilities

The getappniche-com/mcp MCP server includes these tools:

• search_apps: Filter and sort app search results across supported store and market metrics.
• get_app_detail: Retrieve details using an app ID, package name, Apple ID, or store URL.
• get_app_historicals: Return time series for reviews, ratings, downloads, or revenue.
• get_keyword_difficulty: Measure keyword popularity, difficulty, traffic, and opportunity.
• batch_keyword_difficulty: Score up to 10 keywords and sort them by opportunity.
• get_app_reviews: Retrieve live reviews or enriched reviews for monitored workspace apps.
• get_supported_countries: List valid country codes for store queries.

Most operations consume credits. App search, details, historicals, and reviews generally charge based on returned records, while keyword scoring charges per keyword. Country-code lookup is free. Results include a credits_charged value.

Limitations and notes

A GetAppNiche plan and API key are required for paid data operations. The documented rate limit is 60 requests per minute per key. Search and review calls return 50 records by default and allow up to 100; review charges are based on the number of reviews returned. Keyword batches support no more than 10 keywords per call, with credits charged for each keyword.

The repository is a transport bridge rather than a separate data service: the actual data endpoint is hosted by GetAppNiche. OAuth support for clients that cannot send request headers is still in progress, while ChatGPT API-key authentication is documented as available.