# Measure [Health: Active]

**Category:** 📊 Monitoring  
**Repository:** https://github.com/measure-sh/measure  
**GitHub Stars:** 1450  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/measure

## Description
Get to the root cause of mobile app crashes, errors & slow traces with Measure MCP

## Claude Desktop Quick Installation
No generated config for this server: it installs through its own setup steps.

The project's own setup steps (from https://measure.sh/docs/mcp), copied verbatim. Run them in order; they are not an mcpServers config.

1. Add to Claude Code:

```sh
claude mcp add --transport http measure https://agent.measure.sh/mcp
```

## Documentation

## What Measure MCP server does

Measure MCP server brings Measure’s mobile application observability data into an MCP-compatible coding agent. It is intended for investigating production issues involving crashes, application-not-responding events, errors, sessions, and performance traces without switching manually between an agent and the Measure interface.

The underlying Measure platform also covers session replays, app health, bug reports, user journeys, funnels, and network performance. The MCP-facing agent experience is described as providing context about crashes, errors, sessions, and traces so an agent can help analyze mobile issues in natural language.

Measure supports mobile applications built with Android, iOS, Flutter, and React Native. Reach for this server when an agent needs observability context to help trace a mobile problem back to its likely source.

## How it works

Clients connect to the hosted MCP endpoint at `https://agent.measure.sh/mcp`. The Measure Agent then makes relevant Measure context available through MCP for investigation from a coding agent or other compatible client.

Measure data comes from applications instrumented with the Measure SDK. The platform supports monitoring and debugging workflows around crashes, ANRs, traces, sessions, network activity, and user actions. Session replays can include tracked clicks, navigation, and HTTP calls, while bug reports preserve the user-action history associated with an issue.

## Setup and configuration

Use the hosted Measure service by signing up for Measure Cloud and connecting an MCP-compatible client to the hosted endpoint:

```text
https://agent.measure.sh/mcp
```

Before the agent can provide application context, the mobile application must be integrated with the Measure SDK. Measure offers SDK support for Android, iOS, Flutter, and React Native.

Organizations that do not want to use Measure Cloud can self-host Measure on their own infrastructure. The project also provides separate documentation for SDK integration and hosting. Configuration details depend on the selected deployment and client.

## Tools and capabilities

- Investigate mobile crashes and ANRs.
- Examine application errors and performance traces.
- Query session context while debugging an issue.
- Use natural-language requests from a coding agent.
- Analyze production observability data from Measure Cloud or a self-hosted deployment.

## Limitations and notes

The MCP experience depends on data collected by an instrumented application. Without Measure SDK integration, an agent has no application telemetry to investigate. Measure is focused on mobile observability; its supported application platforms are Android, iOS, Flutter, and React Native.

The broader Measure platform includes features such as adaptive capture, session replay, funnels, user journeys, and network performance monitoring. The MCP server is presented as the agent interface for investigating Measure context rather than as a replacement for the full Measure dashboard.

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

