One-click install
npx skills add https://github.com/xz1220/oh-my-kimi --skill debug-xz1220
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/xz1220/oh-my-kimi/tree/main/skills/debug
Command: npx skills add https://github.com/xz1220/oh-my-kimi --skill debug-xz1220

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When a Kimi CLI session, workflow, or hook behaves unexpectedly, this Skill helps you diagnose the issue quickly by locating the actual failure signals rather than guessing.

Core Features & Use Cases

  • Evidence-driven debugging: Focuses on trace, state, and related local records to separate symptoms from root causes.
  • Targeted reproduction: Runs narrow-scope replays when possible to confirm hypotheses safely.
  • Actionable next steps: Produces a minimal next fix or verification step aligned to the evidence found.

Quick Start

Ask the skill to debug your current failure by using trace and state evidence and then propose the smallest next correction step.

Frequently Asked Questions about debug

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug a Kimi CLI workflow failure using trace evidence?

To debug a Kimi CLI workflow failure, prioritize trace and state artifacts to separate symptoms from root causes, then recommend a minimal next fix or verification action without broad rewrites.

What is evidence-based root-cause analysis for runtime session issues?

Evidence-based root-cause analysis for runtime session issues investigates local records and state to distinguish symptoms from actual failure signals, avoiding guesses and targeting the true origin.

How do I troubleshoot hook interactions causing unexpected execution behavior?

Troubleshoot hook interactions causing unexpected execution behavior by inspecting trace and state artifacts to locate actual failure signals, confirming hypotheses with narrow-scope replays before applying fixes.

What is the best way to diagnose orchestration failures without rewriting code?

The best way to diagnose orchestration failures without rewriting code is to locate failure signals in state artifacts, distinguish them from symptoms, and propose the smallest next correction step.

Why does targeted reproduction help with incident triage in active projects?

Targeted reproduction helps with incident triage by running narrow-scope replays to safely confirm hypotheses, ensuring the recommended verification action aligns directly with the evidence found.

Can I use state inspection to find confusing execution behavior in my CLI?

Yes, state inspection locates actual failure signals in your CLI, separating confusing execution symptoms from root causes to recommend a minimal fix without requiring broad rewrites.