Agent skill
debugger
Systematic debugging methodology for diagnosing failures and root cause analysis. Triggers on: 2+ failed fix attempts, 'ultradebug', 'uld', debugging in circles, complex system failures, intermittent bugs.
Install this agent skill to your Project
npx add-skill https://github.com/TechDufus/oh-my-claude/tree/main/plugins/oh-my-claude/skills/debugger
SKILL.md
Debugger Skill
Systematic debugging methodology for diagnosing failures and root cause analysis.
When to Apply This Guidance
- 2+ failed fix attempts on the same issue
- Debugging that's gone in circles
- Complex system behavior you don't understand
- Multi-system integration problems
- Intermittent failures (Heisenbugs)
Structured Analysis Framework
Step 1: Restate the Problem
Before investigating, articulate:
- What is the actual vs expected behavior?
- What assumptions might be wrong?
- What could cause the specific symptoms described?
- Are there hidden dependencies or side effects?
Step 2: Generate Ranked Hypotheses
Always generate multiple hypotheses, ranked by likelihood:
Most Likely:
- Evidence: What supports this theory?
- Test: How to verify quickly?
Possible:
- Evidence: What partially supports this?
- Test: How to confirm or eliminate?
Unlikely but worth checking:
- Evidence: Limited, but possible
- Test: Quick check to rule out
Step 3: Systematic Verification
Test hypotheses in order. For each:
- State what you're testing
- Describe the expected outcome if hypothesis is correct
- Execute the test
- Document actual results
- Update hypothesis ranking based on findings
Reasoning Principles
- Challenge assumptions - The "obvious" cause is often wrong after 2+ failures
- Follow the data - What do logs/errors actually say vs what's assumed?
- Consider timing - Race conditions, async issues, initialization order
- Check boundaries - Module interfaces, API contracts, type conversions
- Question the environment - Config, dependencies, network, state
Decision Framework
When choosing between approaches:
| Factor | Weight | Considerations |
|---|---|---|
| Correctness | Critical | Does it actually solve the problem? |
| Simplicity | High | Prefer boring solutions over clever ones |
| Leverage | High | Use existing patterns/libraries/code |
| Developer Experience | Medium | How hard is it to debug/maintain? |
| Performance | Low* | *Unless performance IS the problem |
Effort Estimation
Tag recommendations by effort:
| Tag | Meaning | Example |
|---|---|---|
| [Quick] | <30 min | Add logging, check config |
| [Short] | 30min-2hr | Refactor function, add test |
| [Medium] | 2hr-1day | New component, integration |
| [Large] | >1 day | Architecture change |
Common Failure Patterns
| Pattern | Signal | Investigation Focus |
|---|---|---|
| Fix-break cycle | Each fix causes new problem | Root cause analysis - treating symptoms not cause |
| Confusion | "I don't understand why X happens" | Add logging, trace execution path |
| Tradeoff paralysis | Multiple valid approaches | Use decision framework above |
| Integration hell | System A + B don't work together | Interface contracts, data formats |
| Heisenbug | Works sometimes, fails randomly | Timing, state, race conditions |
Traps to Avoid
- Premature fixing - Understand before changing
- Single hypothesis - Always consider alternatives
- Assuming correctness - Verify each component independently
- Ignoring evidence - If data contradicts theory, update theory
- Scope creep - Fix the bug, not everything around it
Output Template
When debugging, structure findings as:
## Problem Understanding
{Restate in own words. Challenge framing if wrong.}
## Hypotheses (Ranked)
1. Most Likely: {hypothesis} - Evidence: {support} - Test: {verification}
2. Possible: {hypothesis} - Evidence: {support} - Test: {verification}
3. Unlikely: {hypothesis} - Evidence: {limited} - Test: {quick check}
## Recommended Approach
{Specific next steps. Be prescriptive.}
## Traps to Avoid
{What NOT to do for this specific problem.}
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
tdd
Systematic TDD methodology for writing tests first. Use when implementing features or fixes with OMC_TDD_MODE enabled, when unsure how to structure tests, or when test-first discipline needs reinforcement. Triggers on: 'tdd', 'test first', 'test driven', 'red green refactor'.
pr-creation
MUST be used when creating pull requests. Handles context gathering, title generation (conventional commit format), body formatting, and PR creation via GitHub CLI. Creates PRs as drafts by default. Triggers on: 'create PR', 'open PR', 'ready for review', 'push for PR', 'send for review'.
ralph-loop-init
Transform approved plans into ralph loop infrastructure. Triggers on: '/ralph-loop-init', '/ralph-init', 'setup ralph loop', 'generate ralph loop'. Creates .ralph/ directory with prd.json, loop.py, CLAUDE.md, and supporting files.
writing-skills
Methodology for creating effective skills using TDD principles. Use when creating new skills, editing existing skills, or authoring plugin components. Triggers on: 'create skill', 'write skill', 'new skill', 'skill authoring'.
ralph-plan
Structured PRD generation with interview, research, and approval workflow. Triggers on: '/ralph-plan <topic>', 'create prd', 'generate prd', 'plan this'. Creates comprehensive Product Requirements Document via interview and research.
verification
Evidence-based verification methodology before claiming work is complete. Use when about to claim work is done, before committing, before creating PRs, or when verification-reminder hook fires. Triggers on: 'verify', 'verification', 'is it done', 'complete', 'ready to merge'.
Didn't find tool you were looking for?