Trace

When you hit a bug or performance issue and don't know why, this skill orchestrates parallel investigation lanes instead of letting you chase the first theory that sounds good. It spins up three tracer workers in Claude's team mode, each pursuing a deliberately different hypothesis: code path causes, config and environment issues, or measurement and assumption mismatches. Each worker gathers evidence for and against their theory, then they run a rebuttal round before ranking explanations. The output gives you the most likely cause, the critical unknown keeping competing theories alive, and the single best probe to collapse uncertainty. Treats tracing as a discipline, not just debugger output with commentary.

skills.sh installs: 840
GitHub stars: 40.7K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add yeachan-heo/oh-my-claudecode --skill trace --agent claude-code

Installs into .claude/skills of the current project.

Files

Trace Skill

Use this skill for ambiguous, causal, evidence-heavy questions where the goal is to explain why an observed result happened, not to jump directly into fixing or rewriting code.

This is the orchestration layer on top of the built-in tracer agent. The goal is to make tracing feel like a reusable OMC operating lane: restate the observation, generate competing explanations, gather evidence in parallel, rank the explanations, and propose the next probe that would collapse uncertainty fastest.