# On-Call Health

> On-Call Health is an open-source MCP server for health and science. Burnout risk analysis for on-call engineering teams, queried from an AI assistant. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI
- **Website:** https://oncallhealth.ai
- **Repository:** https://github.com/rootly-ai-labs/mcp-on-call-health
- **MCP Registry name:** io.github.Rootly-AI-Labs/oncallhealth-mcp
- **Categories:** health and science, developer tools, analytics and data
- **Listing updated:** 2026-11-14

## About

The server turns On-Call Health data into answers about workload: team risk summaries, which responders are at risk, and daily health trends. It runs locally over stdio, started with uvx from the PyPI package oncallhealth-mcp, and needs an API key from oncallhealth.ai settings in the ONCALLHEALTH_API_KEY variable.

The README suggests a workflow. Begin with validate_integrations to confirm the data sources are connected, then analysis_current for the latest analysis or analysis_run to start a fresh one, then analysis_summary and the member-level tools. While a run is in progress, the server polls every five seconds by default, backing off to thirty. An optional Code Mode exposes the tool catalogue through meta-tools instead of listing every tool.

It implements MCP revision 2026-07-28 and still negotiates older revisions. Every tool publishes read-only, idempotent and open-world annotations plus a structured output schema. The product site describes an open source tool that looks for early signs of overload in on-call engineers, from Rootly AI Labs.

## Features

- **Team risk summaries:** Overview of burnout risk across an on-call team.
- **At-risk responders:** Identify members showing strong overload signals.
- **Daily health trends:** Track how team health changes day by day.
- **Fresh analyses:** analysis_run starts a new analysis and polls until it finishes.
- **Integration check:** validate_integrations confirms connected sources first.
- **Tool annotations:** Every tool declares read-only, idempotent and structured output.
- **Local stdio server:** Runs with uvx and a single API key.

## Good for

- Check which responders are at risk this week
- Compare team health day by day after a bad incident
- Run a new burnout analysis before a rota change
- Summarise on-call load in a retro

## Install

### Claude Code (Terminal)

```sh
claude mcp add oncallhealth-mcp --env ONCALLHEALTH_API_KEY='<ONCALLHEALTH_API_KEY>' -- uvx oncallhealth-mcp
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "oncallhealth-mcp": {
      "command": "uvx",
      "args": [
        "oncallhealth-mcp"
      ],
      "env": {
        "ONCALLHEALTH_API_KEY": "<ONCALLHEALTH_API_KEY>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "oncallhealth-mcp": {
      "command": "uvx",
      "args": [
        "oncallhealth-mcp"
      ],
      "env": {
        "ONCALLHEALTH_API_KEY": "<ONCALLHEALTH_API_KEY>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "oncallhealth-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "oncallhealth-mcp"
      ],
      "env": {
        "ONCALLHEALTH_API_KEY": "<ONCALLHEALTH_API_KEY>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "oncallhealth-mcp": {
      "command": "uvx",
      "args": [
        "oncallhealth-mcp"
      ],
      "env": {
        "ONCALLHEALTH_API_KEY": "<ONCALLHEALTH_API_KEY>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "oncallhealth-mcp": {
      "command": "uvx",
      "args": [
        "oncallhealth-mcp"
      ],
      "env": {
        "ONCALLHEALTH_API_KEY": "<ONCALLHEALTH_API_KEY>"
      }
    }
  }
}
```

## Tools

- `validate_integrations`: Checks that the connected data sources are working.
- `analysis_current`: Returns the current analysis.
- `analysis_run`: Starts a fresh analysis and polls until it completes.
- `analysis_summary`: Returns a summary of an analysis.

## FAQ

### What do I need to use it?

An On-Call Health account and an API key from oncallhealth.ai/settings/api-keys.

### How is it installed?

Run it with uvx oncallhealth-mcp, or pip install oncallhealth-mcp, with ONCALLHEALTH_API_KEY set.

### Which tool should I call first?

validate_integrations, then analysis_current or analysis_run.

### Which clients are documented?

Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, Codex CLI and Gemini CLI.

### Can I point it at another API URL?

Yes, set ONCALLHEALTH_API_URL; the default is https://api.oncallhealth.ai.

### Is On-Call Health free?

Yes. On-Call Health is open source, so you can use and self-host it for free.

### How do I add On-Call Health to Claude Code?

Run this command in your terminal: claude mcp add oncallhealth-mcp --env ONCALLHEALTH_API_KEY='<ONCALLHEALTH_API_KEY>' -- uvx oncallhealth-mcp

### What are the best alternatives to On-Call Health?

The closest alternatives to On-Call Health are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/on-call-health
