Agent Test Long Runner

This is basically a test harness for validating Claude's ability to handle tasks that run 30+ minutes without timing out or losing context. It's designed to do comprehensive work like full codebase analysis, architecture design docs, or security audits where you need thorough results rather than quick answers. The agent explicitly prioritizes quality over speed and includes progress communication so you're not sitting there wondering if it crashed. Honestly most useful for QA teams testing agent reliability or developers who need to validate that their complex, multi-step workflows won't fall apart halfway through.

Install to Claude Code

npx -y skills add ruvnet/ruflo --skill agent-test-long-runner --agent claude-code