VividKit Crystal clear AI coding
VI
Choose Your Setup Get Started with AgentKit Continue with ClaudeKit Migrate from ClaudeKit to AgentKit Clean cutover from ClaudeKit What Is AgentKit? How AgentKit Works AgentKit Kits, Skills, Agents & Hooks AgentKit Runtime Support AgentKit Desktop App AgentKit Configuration AgentKit CLI Quick Reference AgentKit Skills Cheatsheet AgentKit Workflows AgentKit Statusline Update AgentKit CLI & Kits AgentKit Helper Clean reinstall of AgentKit AgentKit Troubleshooting What is ClaudeKit? ClaudeKit CLI Guide Command Finder Commands Cheat Sheet Workflow Recipes CLI Quick Reference Claude Mechanics Permission Modes ClaudeKit Hooks Coexistence ClaudeKit x Codex CK → Other coding tools How ClaudeKit Works Inside ClaudeKit Finding Your Unknowns Session Recovery Fix from Logs UI/UX Guide Accounts Switcher IDE Config Remote Control Codex App System Diagramming UI Review Gate Promotions Support VividKit

Quick Ref / Command

Marketing Kit

/ak:debugging

Root-cause debugging discipline

Forces root-cause investigation, backward tracing, defense-in-depth validation, and fresh verification before claiming bugs, tests, builds, or performance issues are fixed.

01

Read the failure

02

Investigate root cause

03

Compare patterns

04

Trace backward

Rule 01

Random fixes waste time and create new bugs.

Rule 02

Fix at the source of invalid data or behavior, not at the visible symptom.

Rule 03

Completion requires fresh verification evidence that you have actually read.

Execution Map

Core lanes from input to output, aligned with the command's real execution path.

01

Start

Input, scope, route

  1. 1 Read the failure Collect the actual error, reproduction, recent changes, and evidence before touching code.
  2. 2 Investigate root cause Complete systematic debugging Phase 1 before any fix attempt.
02

Work

Agent / skill execution

  1. 3 Compare patterns Find working examples, compare differences, and identify why this path fails.
  2. 4 Trace backward If the error appears deep in execution, walk up the call stack until the invalid data or trigger originates.
03

Verify

Gate, review, validation

  1. 5 Test hypothesis Form a minimal theory and test it before implementing the final fix.
  2. 6 Fix at source Repair the original cause, not just the symptom where it surfaced.
04

Close

Report, handoff, artifact

  1. 7 Add depth After root cause is known, add entry validation, business logic guards, environment guards, or debug instrumentation as appropriate.
  2. 8 Verify fresh Run the verification command, read the output, and only then claim success.

Don't skip the required step

Each item contrasts a tempting shortcut with the required approach.

Tempting shortcut

It is probably X; I can fix that now.

Do this instead

That is a red flag; return to systematic investigation.

Tempting shortcut

Should work now.

Do this instead

Run verification and read the output before claiming it works.

Sample Prompt

Concrete invocations for each flag, subcommand, mode, or route available in the live workflow.

Bug investigation Recommended 
/ak:debugging checkout submit fails with 500
Use when:
Use when behavior is broken and the cause is not proven.
Expected:
A root-cause-first investigation and source fix path.
Test failure 
/ak:debugging failing auth middleware tests
Use when:
Use when a test failure needs evidence before code changes.
Expected:
Failure reading, pattern comparison, hypothesis testing, and fresh verification.
Performance issue 
/ak:debugging dashboard loads slowly
Use when:
Use when performance is unexpectedly poor.
Expected:
Evidence-based tracing to the real bottleneck before optimization.

Handled Scope

  • Systematic debugging
  • Root-cause tracing
  • Defense-in-depth
  • Verification protocols
  • Test failures
  • Performance issues
  • Build failures

Next