# ARPI MCP

**Category:** 🧬 Biology & Bioinformatics  
**Repository:** https://github.com/arpi-ai/arpi-mcp  
**GitHub Stars:** 0  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/arpi-mcp

## Description
AI-powered ECG analysis: submit ECG images and receive diagnosis reports from ARPI's ECG AI.

## Tools
Capabilities this server exposes over MCP:

- **authenticate** — Sign in to ARPI
- **open_image_submission** — Start an ECG image submission
- **submit_image_for_diagnosis** — Submit an ECG image for AI diagnosis
- **fetch_diagnosis_reports** — Retrieve diagnosis reports
- **list_available_capabilities** — List available analysis capabilities
- **invoke_arpi_capability** — Invoke a specific analysis capability

## 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": {
  "arpi-mcp": {
    "url": "https://mcp.arpi.ai/mcp"
  }
}
```

## Documentation

## What the ARPI MCP MCP server does

The ARPI MCP MCP server exposes ARPI’s ECG AI through the Model Context Protocol. An MCP client can authenticate with ARPI, begin an ECG image submission, send the image for analysis, and retrieve generated diagnosis reports. The service also exposes capability discovery and a general operation for invoking a named ARPI analysis capability.

This is a remote service rather than a server that runs from the repository. The repository contains registry and marketplace metadata, including the official MCP Registry entry and Claude Code plugin files. The hosted MCP endpoint is operated by ARPI at `https://mcp.arpi.ai/mcp`.

## How it works

The ARPI MCP MCP server communicates over Streamable HTTP. Authentication uses OAuth, and the MCP client guides the user through ARPI login during the first connection. After authentication, the client can use the image-submission workflow and retrieve reports through the exposed tools.

The typical workflow is to start an image submission, submit an ECG image for diagnosis, and fetch the resulting reports. Clients can also query the analysis capabilities available to the authenticated account and invoke a specific capability when that is appropriate.

## Setup and configuration

No local package or process is described for the MCP server itself. Configure an MCP client to use this endpoint:

```text
https://mcp.arpi.ai/mcp
```

Claude Code can register it with the HTTP transport using:

```bash
claude mcp add --transport http arpi-mcp https://mcp.arpi.ai/mcp
```

The repository also provides a Claude Code marketplace option. Add the `arpi-ai/arpi-mcp` marketplace and install the `arpi-mcp@arpi-mcp` plugin. In Claude.ai or Claude Desktop, add a custom connector with the same endpoint. Other clients need to support both Streamable HTTP and OAuth.

## Tools and capabilities

The available tools are:

- `authenticate`: Sign in to ARPI.
- `open_image_submission`: Begin an ECG image submission.
- `submit_image_for_diagnosis`: Send an ECG image for AI diagnosis.
- `fetch_diagnosis_reports`: Retrieve diagnosis reports.
- `list_available_capabilities`: View available analysis capabilities.
- `invoke_arpi_capability`: Invoke a selected ARPI capability.

Together, these tools cover authentication, ECG image analysis, report retrieval, and access to ARPI’s advertised capabilities. The material does not specify image formats, report schemas, response times, or the names of individual capabilities.

## Limitations and notes

The ARPI MCP MCP server connects to a medical AI service. Its diagnosis reports are intended for reference and must not be treated as a replacement for professional medical judgment.

Because the service is remote, availability and authentication depend on ARPI’s hosted endpoint and OAuth flow. Support is provided by ARPI at `support@arpi.ai`.

_Full upstream README: https://allmcps.com/mcp/arpi-mcp/readme_