# AsoBeast [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/AsoBeast/asobeast  
**GitHub Stars:** 49  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/asobeast

## Description
App Store and Google Play ASO: keyword rankings, competitors, reviews, audits and an action queue.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "asobeast": {
    "url": "https://docs.asobeast.com/mcp/introduction"
  }
}
```

## Documentation

## What AsoBeast MCP server does

The AsoBeast MCP server gives AI clients access to an AsoBeast ASO workspace covering Apple App Store and Google Play listings. It is intended for working with app visibility data rather than general software development tasks. An instance can contain imported listings, tracked keywords, storefront markets, competitor data, reviews, metadata history, audits, and ASO recommendations.

The MCP interface includes 27 read-only tools. These let an agent inspect information already collected by AsoBeast, including keyword positions and history, competitors, reviews, listing metadata, SERP snapshots, category data, audits, and action recommendations. Five additional tools are opt-in and can change keywords, competitors, or actions.

## How it works

AsoBeast collects store information from public Apple and Google requests made by the machine hosting the application. Keyword checks can track positions to a depth of 200, while one search serves the app and its competitors. The resulting history supports ranking analysis, competitor comparisons, review and metadata change detection, audits, and a deterministic action queue.

The AsoBeast MCP server exposes that stored data to an MCP client through stdio or a remote endpoint. Read tools are available for inspection; tools that modify keywords, competitors, or actions require explicit opt-in. Recommendations are generated by fifteen deterministic rules, with supporting evidence, priority, estimated impact, lifecycle state, and verified outcome information.

## Setup and configuration

The open-source edition can be self-hosted with Docker Compose. The documented local setup requires Docker, about 2 GB of memory, and about 5 GB of disk. The standard deployment uses PostgreSQL and Redis, creates an owner account through the web interface, and serves the application locally on port 3001. Published container images can also be used without cloning the repository, and a Railway deployment option is provided.

The Docker setup requires `POSTGRES_PASSWORD` and `AUTH_SECRET` in the environment file. AsoBeast does not require an ASO vendor account or vendor API key for its store collection. Optional outbound features can send data through webhooks, SMTP email, OpenAI assistance, Apple Ads search popularity, or store status updates, but those integrations are outside the core MCP connection.

## Tools and capabilities

The AsoBeast MCP server supports agent workflows such as:

- Reviewing current and historical keyword rankings across storefronts.
- Comparing app listings and discovering competitors from search results.
- Inspecting review language, rating history, and metadata changes.
- Reading audit results, metadata lints, keyword groups, and SERP movement.
- Examining prioritized ASO actions and their supporting evidence.
- Updating keywords, competitors, or actions through the opt-in mutation tools.

The underlying application also supports portfolio grouping across stores and countries, weekly digests, alerts, and metadata workbench features. MCP access is described as 27 read tools and five opt-in mutation tools; the excerpt does not provide individual tool names or schemas.

## Limitations and notes

The AsoBeast MCP server requires an AsoBeast instance and does not independently provide a store-data service. Its available context depends on what the connected instance has imported and tracked. Store collection is performed at a deliberately modest rate, and keyword checks represent public store searches rather than a private ranking feed.

Only the five opt-in tools change keywords, competitors, or actions; the remaining tools are for reading data. The README names Claude Code and Claude Desktop as clients that can work with the instance. It does not document client-specific setup details or the individual MCP tool schemas in the provided material.

_Full upstream README: https://allmcps.com/mcp/asobeast/readme_