alibaba/arthas

Arthas is a Java diagnostic tool that provides a comprehensive suite of runtime analysis capabilities, including deep classloader inspection, memory profiling, and JVM monitoring. The system features a modular architecture supporting interactive shell, HTTP/WebSocket APIs, and an AI-driven Model Context Protocol (MCP) interface for automated diagnostics. It enables developers to execute commands for tracing, logging, and heap analysis, while also offering a web-based console for remote management and agent discovery.

This period focused on establishing the foundational architecture for the Arthas shell, introducing a new session management system, command execution framework, and terminal abstraction. The codebase was also restructured into a multi-module Maven project, extracting common utilities and separating the Java-based bootstrapper from shell scripts.

42 changes

This period focused on enabling remote agent management through a new tunnel client and server architecture, supporting WebSocket-based connections and cluster state persistence. Concurrently, the project expanded its HTTP-based terminal and API capabilities, introducing a unified HTTP/Telnet server, structured command result models, and a Spring Boot starter for seamless integration.

34 changes

This period focused on expanding Arthas's capabilities through the introduction of the Model Context Protocol (MCP) server, enabling AI-driven diagnostics and tool integration. Concurrently, the project significantly deepened its native agent support, adding low-level JVM memory analysis, heap inspection, and web UI management for native processes. The work also included refactoring core models and adding integration tests to ensure stability.

22 changes

alibaba/arthas was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point the surveyor at a repository you know and see whether you agree with it.