Systematic Debugging
Random fixes waste time and create new bugs. Quick patches mask underlying issues.
Core principle: ALWAYS find root cause before attempting fixes.
Usage
/debug
Invoke this skill when encountering any bug, test failure, or unexpected behavior. Follow the four phases in order.
When to use
- Test failures
- Bugs in production
- Unexpected behavior
- Performance problems
- Build failures
- Integration issues
Use this ESPECIALLY when:
- Under time pressure (emergencies make guessing tempting)
- "Just one quick fix" seems obvious
- You've already tried multiple fixes
- Previous fix didn't work
The Iron Law
NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST
If you haven't completed Phase 1, you cannot propose fixes.
The Four Phases
Complete each phase before proceeding to the next.
Phase 1: Root Cause Investigation
BEFORE attempting ANY fix:
Read Error Messages Carefully
- Don't skip past errors or warnings
- Read stack traces completely
- Note line numbers, file paths, error codes
Reproduce Consistently
- Can you trigger it reliably?
- What are the exact steps?
- If not reproducible, gather more data — don't guess
Check Recent Changes
- What changed that could cause this?
- Git diff, recent commits
- New dependencies, config changes
Gather Evidence in Multi-Component Systems
- Log what data enters each component boundary
- Log what data exits each component boundary
- Verify environment/config propagation
- Run once to gather evidence showing WHERE it breaks
- THEN investigate that specific component
Trace Data Flow
- Where does the bad value originate?
- What called this with the bad value?
- Keep tracing up until you find the source
- Fix at source, not at symptom
Phase 2: Pattern Analysis
- Find Working Examples — locate similar working code in the same codebase
- Compare Against References — read reference implementations COMPLETELY, don't skim
- Identify Differences — list every difference between working and broken, however small
- Understand Dependencies — what components, settings, environment does it need?
Phase 3: Hypothesis and Testing
- Form Single Hypothesis — "I think X is the root cause because Y"
- Test Minimally — smallest possible change, one variable at a time
- Verify — did it work? Yes = Phase 4. No = new hypothesis. DON'T stack fixes.
Phase 4: Implementation
- Create Failing Test Case — simplest possible reproduction, automated if possible. Use
/tdd for proper failing tests.
- Implement Single Fix — address the root cause, ONE change at a time, no "while I'm here" improvements.
- Verify Fix — test passes? No other tests broken? Issue actually resolved? Use
/verify before claiming done.
If fix doesn't work:
- If < 3 attempts: return to Phase 1, re-analyze with new information
- If >= 3 attempts: STOP and question the architecture
3+ failed fixes indicate an architectural problem:
- Each fix reveals new coupling in different places
- Fixes require massive refactoring
- Each fix creates new symptoms elsewhere
- Discuss with the user before attempting more fixes
Red Flags - STOP and Follow Process
- "Quick fix for now, investigate later"
- "Just try changing X and see if it works"
- "Add multiple changes, run tests"
- "It's probably X, let me fix that"
- "I don't fully understand but this might work"
- Proposing solutions before tracing data flow
- "One more fix attempt" (when already tried 2+)
ALL of these mean: STOP. Return to Phase 1.
Common Rationalizations
| Excuse |
Reality |
| "Issue is simple, don't need process" |
Simple issues have root causes too. Process is fast for simple bugs. |
| "Emergency, no time for process" |
Systematic debugging is FASTER than guess-and-check thrashing. |
| "Just try this first, then investigate" |
First fix sets the pattern. Do it right from the start. |
| "I'll write test after confirming fix" |
Untested fixes don't stick. Test first proves it. |
| "Multiple fixes at once saves time" |
Can't isolate what worked. Causes new bugs. |
| "I see the problem, let me fix it" |
Seeing symptoms is not understanding root cause. |
Quick Reference
| Phase |
Key Activities |
Success Criteria |
| 1. Root Cause |
Read errors, reproduce, check changes, gather evidence |
Understand WHAT and WHY |
| 2. Pattern |
Find working examples, compare |
Identify differences |
| 3. Hypothesis |
Form theory, test minimally |
Confirmed or new hypothesis |
| 4. Implementation |
Create test, fix, verify |
Bug resolved, tests pass |
Related Skills
/tdd — TDD cycle for creating failing test cases in Phase 4
/verify — verification before claiming the fix is complete
1---2name: debug3description: "Discipline skill — systematic debugging process. Use when encountering any bug, debug issue, test failure, broken function, or unexpected behavior, before proposing fixes."4user-invocable: true5metadata:6 source: https://skillmd.com/skills/guild-agents/debug7---89# Systematic Debugging1011Random fixes waste time and create new bugs. Quick patches mask underlying issues.1213**Core principle:** ALWAYS find root cause before attempting fixes.1415## Usage1617`/debug`1819Invoke this skill when encountering any bug, test failure, or unexpected behavior. Follow the four phases in order.2021## When to use2223- Test failures24- Bugs in production25- Unexpected behavior26- Performance problems27- Build failures28- Integration issues2930**Use this ESPECIALLY when:**3132- Under time pressure (emergencies make guessing tempting)33- "Just one quick fix" seems obvious34- You've already tried multiple fixes35- Previous fix didn't work3637## The Iron Law3839```text40NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST41```4243If you haven't completed Phase 1, you cannot propose fixes.4445## The Four Phases4647Complete each phase before proceeding to the next.4849### Phase 1: Root Cause Investigation5051**BEFORE attempting ANY fix:**52531. **Read Error Messages Carefully**54 - Don't skip past errors or warnings55 - Read stack traces completely56 - Note line numbers, file paths, error codes57582. **Reproduce Consistently**59 - Can you trigger it reliably?60 - What are the exact steps?61 - If not reproducible, gather more data — don't guess62633. **Check Recent Changes**64 - What changed that could cause this?65 - Git diff, recent commits66 - New dependencies, config changes67684. **Gather Evidence in Multi-Component Systems**69 - Log what data enters each component boundary70 - Log what data exits each component boundary71 - Verify environment/config propagation72 - Run once to gather evidence showing WHERE it breaks73 - THEN investigate that specific component74755. **Trace Data Flow**76 - Where does the bad value originate?77 - What called this with the bad value?78 - Keep tracing up until you find the source79 - Fix at source, not at symptom8081### Phase 2: Pattern Analysis82831. **Find Working Examples** — locate similar working code in the same codebase842. **Compare Against References** — read reference implementations COMPLETELY, don't skim853. **Identify Differences** — list every difference between working and broken, however small864. **Understand Dependencies** — what components, settings, environment does it need?8788### Phase 3: Hypothesis and Testing89901. **Form Single Hypothesis** — "I think X is the root cause because Y"912. **Test Minimally** — smallest possible change, one variable at a time923. **Verify** — did it work? Yes = Phase 4. No = new hypothesis. DON'T stack fixes.9394### Phase 4: Implementation95961. **Create Failing Test Case** — simplest possible reproduction, automated if possible. Use `/tdd` for proper failing tests.972. **Implement Single Fix** — address the root cause, ONE change at a time, no "while I'm here" improvements.983. **Verify Fix** — test passes? No other tests broken? Issue actually resolved? Use `/verify` before claiming done.99100**If fix doesn't work:**101102- If < 3 attempts: return to Phase 1, re-analyze with new information103- **If >= 3 attempts: STOP and question the architecture**104105**3+ failed fixes indicate an architectural problem:**106107- Each fix reveals new coupling in different places108- Fixes require massive refactoring109- Each fix creates new symptoms elsewhere110- Discuss with the user before attempting more fixes111112## Red Flags - STOP and Follow Process113114- "Quick fix for now, investigate later"115- "Just try changing X and see if it works"116- "Add multiple changes, run tests"117- "It's probably X, let me fix that"118- "I don't fully understand but this might work"119- Proposing solutions before tracing data flow120- "One more fix attempt" (when already tried 2+)121122**ALL of these mean: STOP. Return to Phase 1.**123124## Common Rationalizations125126| Excuse | Reality |127| ----------------------------------------- | ------------------------------------------------------------------------ |128| "Issue is simple, don't need process" | Simple issues have root causes too. Process is fast for simple bugs. |129| "Emergency, no time for process" | Systematic debugging is FASTER than guess-and-check thrashing. |130| "Just try this first, then investigate" | First fix sets the pattern. Do it right from the start. |131| "I'll write test after confirming fix" | Untested fixes don't stick. Test first proves it. |132| "Multiple fixes at once saves time" | Can't isolate what worked. Causes new bugs. |133| "I see the problem, let me fix it" | Seeing symptoms is not understanding root cause. |134135## Quick Reference136137| Phase | Key Activities | Success Criteria |138| ---------------------- | ------------------------------------------------------ | --------------------------- |139| **1. Root Cause** | Read errors, reproduce, check changes, gather evidence | Understand WHAT and WHY |140| **2. Pattern** | Find working examples, compare | Identify differences |141| **3. Hypothesis** | Form theory, test minimally | Confirmed or new hypothesis |142| **4. Implementation** | Create test, fix, verify | Bug resolved, tests pass |143144## Related Skills145146- `/tdd` — TDD cycle for creating failing test cases in Phase 4147- `/verify` — verification before claiming the fix is complete