Debug

by Yeachan-Heo454bae017356ed1d056182cd929b6eb6cf6143adNo licenseListed Oct 9, 2026Updated Oct 9, 2026

Diagnose the current OMC session or repo state using logs, traces, state, and focused reproduction

AI-generated overview

Diagnoses OMC or Claude Code session failures using logs, traces, state, and narrow reproduction.

What it does: Guides an agent through diagnosing a current OMC or Claude Code session problem, workflow breakage, or confusing runtime behavior. It inspects local evidence such as trace tools, state tools, project memory, and failing tests or commands, then reproduces the issue narrowly. It reports the observed failure, a root-cause hypothesis with supporting evidence, and the smallest next corrective or verification step.
When to use it: Use it when a user asks for help diagnosing a current OMC or Claude Code session problem, a broken workflow, or confusing runtime behavior. It fits situations where orchestration, hooks, or agent flow may be involved and evidence must be separated from symptoms.
Requirements: Instructions only; ships no scripts. It relies on access to local session evidence such as trace and state tools, project memory, and failing tests or commands.

Use this skill when the user wants help diagnosing a current OMC/Claude-Code session problem, workflow breakage, or confusing runtime behavior.

Find the real failure signal quickly and explain the next corrective step.

Read the user’s issue description carefully.
Inspect the most relevant local evidence first:
- trace tools
- state tools
- notepad / project memory when relevant
- failing tests or commands
Reproduce the issue narrowly if possible.
Distinguish symptoms from root cause.
Recommend the smallest next fix or verification step.

Prefer real evidence over guesses.
Use the trace/state surfaces when the issue involves orchestration, hooks, or agent flow.
If the issue is actually a product/runtime bug rather than app code, say so plainly.
Do not prescribe broad rewrites before isolating the failure.
Tag temporary debug output [DEBUG-<4hex>]; cleanup is then a single grep.
Redact secrets from logs and traces before they leave the session.
If the failure traces to a missing or wrong seam, say so as the finding and point at refit or architecture-survey — the environment lesson outlives the fix.

Source and attribution

License: No license

Content belongs to its original authors. SourceWeft indexes it from a public repository.