en un clic
systematic-debugging
Structured approach to finding and fixing bugs
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Menu
Structured approach to finding and fixing bugs
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Create or update Pi skills (SKILL.md plus optional scripts, references, or assets). Use when someone asks to design a new Pi skill, refine an existing one, or structure skills for Pi discovery or packaging.
Guide for extending Pi — decide between skills, extensions, prompt templates, themes, context files, or custom models, then create and package them. Use when someone wants to extend Pi, add capabilities, create a skill, build an extension, or make a Pi package.
Long-running iterative development loops with pacing control and verifiable progress. Use when tasks require multiple iterations, many discrete steps, or periodic reflection with clear checkpoints; avoid for simple one-shot tasks or quick fixes.
Cheat sheet + workflow for launching interactive coding-agent CLIs (Claude Code, Gemini CLI, Codex CLI, Cursor CLI, and pi itself) via the interactive_shell overlay or headless dispatch. Use for TUI agents and long-running processes that need supervision, fire-and-forget delegation, or headless background execution. Regular bash commands should use the bash tool instead.
Clarify requirements before implementing. Do not use automatically, only when invoked explicitly.
Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.
| name | systematic-debugging |
| description | Structured approach to finding and fixing bugs |
| version | 1.0.0 |
| author | Ariff |
| when_to_use | When debugging - follow systematic process instead of guessing |
REPRODUCE → ISOLATE → TRACE → FIX → VERIFY
Never skip steps. Never guess.
Goal: See the bug happen reliably
1. Get exact reproduction steps
2. Run them yourself
3. See the exact error
4. Document: input, output, expected
Red flags:
Goal: Find minimum case that shows bug
1. Start removing things
2. Does bug still happen?
3. Keep removing until minimal
4. Result: smallest repro case
Why this matters:
Goal: Find root cause (use root-cause-tracing skill)
1. Start at symptom
2. Trace backward: where does bad value come from?
3. Keep asking "where did THIS come from?"
4. Stop when you find origin
Tools:
Goal: Change code at root cause
1. Write failing test first
2. Make minimal change to fix
3. Verify test passes
4. Check for side effects
Anti-patterns:
Goal: Confirm fix works and nothing broke
1. Run original reproduction
2. Bug should be gone
3. Run full test suite
4. No regressions
Must have:
| Type | Debug Strategy |
|---|---|
| Null/undefined | Trace where value should've been set |
| Wrong value | Log at each transformation step |
| Race condition | Add logging with timestamps |
| State corruption | Find all state modifiers |
| Off-by-one | Check loop bounds and indices |
| Missing case | Check all switch/if branches |
When adding debug logs:
// Include: location, variable name, value, timestamp
console.log('[user.ts:45]', 'userId:', userId, Date.now());
Clean up logs before committing.
When you don't know what broke it:
1. Find last known working version
2. Find first broken version
3. Binary search between them
4. Each step: does bug exist?
5. Result: exact commit that broke it
root-cause-tracing → For step 3 (Trace)defense-in-depth → For step 5 (Verify)verification-before-completion → Before claiming fixedassumption-checker → What assumptions about the code?fact-checker → Verify claims about behaviorpre-action-verifier → Before applying fix