skill-debug

Host: Codex CLI — This skill was designed for Claude Code and adapted for Codex. Cross-reference commands use installed skill names in Codex rather than /octo:* slash commands. Use the active Codex shell and subagent tools. Do not claim a provider, model, or host subagent is available until the current session exposes it. For host tool equivalents, see skills/blocks/codex-host-adapter.md.

Systematic Debugging

MANDATORY COMPLIANCE — DO NOT SKIP

When this skill is invoked, you MUST follow the 4-phase debugging process below. You are PROHIBITED from:

• Jumping straight to a fix without completing Phase 1 (Root Cause Investigation)
• Skipping the hypothesis step and guessing at solutions
• Deciding the bug is "obvious" and bypassing the systematic process
• Attempting more than 3 fixes without stopping to ask the user

Systematic debugging finds root causes in 15-30 minutes. Random fixes waste 2-3 hours. Follow the process.

Your first output line MUST be: 🐙 **CLAUDE OCTOPUS ACTIVATED** - Systematic Debugging

The Iron Law

NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST

Random fixes waste time and create new bugs. Quick patches mask underlying issues.

If you haven't completed Phase 1, you cannot propose fixes.

When to Use

Use for ANY technical issue:

• Test failures
• Bugs in production
• Unexpected behavior
• Performance problems
• Build failures
• Integration issues

Use 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 Four Phases

┌──────────────────┐
│ Phase 1: ROOT    │ ← Understand WHAT and WHY
│ CAUSE            │
└────────┬─────────┘
         ↓
┌──────────────────┐
│ Phase 2: PATTERN │ ← Find working examples
│ ANALYSIS         │
└────────┬─────────┘
         ↓
┌──────────────────┐
│ Phase 3:         │ ← Form and test hypothesis
│ HYPOTHESIS       │
└────────┬─────────┘
         ↓
┌──────────────────┐
│ Phase 4:         │ ← Fix root cause, not symptom
│ IMPLEMENTATION   │
└──────────────────┘

You MUST complete each phase before proceeding.

Phase 1: Root Cause Investigation

BEFORE attempting ANY fix:

1. Read Error Messages Carefully

• Don't skip past errors or warnings
• Read stack traces completely
• Note line numbers, file paths, error codes
• Error messages often contain the exact solution

2. Reproduce Consistently

• Can you trigger it reliably?
• What are the exact steps?
• Does it happen every time?
• If not reproducible → gather more data, don't guess

3. Check Recent Changes

git diff HEAD~5
git log --oneline -10

• What changed that could cause this?
• New dependencies, config changes?
• Environmental differences?

4. Gather Evidence in Multi-Component Systems

When system has multiple components (API → service → database):

# Add diagnostic instrumentation at EACH boundary
echo "=== Layer 1: API endpoint ==="
echo "Input: $INPUT"

echo "=== Layer 2: Service layer ==="
echo "Received: $DATA"

echo "=== Layer 3: Database ==="
echo "Query: $QUERY"

Run once to gather evidence showing WHERE it breaks.

5. Trace Data Flow

When error is deep in call stack:

• Where does bad value originate?
• What called this with bad value?
• Keep tracing up until you find the source
• Fix at source, not at symptom

Phase 2: Pattern Analysis

1. Find Working Examples

• Locate similar working code in same codebase
• What works that's similar to what's broken?

2. Compare Against References

• If implementing a pattern, read reference implementation COMPLETELY
• Don't skim - read every line
• Understand the pattern fully before applying

3. Identify Differences

• What's different between working and broken?
• List every difference, however small
• Don't assume "that can't matter"

4. Understand Dependencies

• What other components does this need?
• What settings, config, environment?
• What assumptions does it make?

Phase 3: Hypothesis and Testing

1. Form Single Hypothesis

• State clearly: "I think X is the root cause because Y"
• Write it down
• Be specific, not vague

2. Test Minimally

• Make the SMALLEST possible change to test hypothesis
• One variable at a time
• Don't fix multiple things at once

3. Verify Before Continuing

Result	Action
Hypothesis confirmed	Proceed to Phase 4
Hypothesis wrong	Form NEW hypothesis, return to Phase 3.1
Still unclear	Gather more evidence, return to Phase 1

4. When You Don't Know

• Say "I don't understand X"
• Don't pretend to know
• Ask for help or research more

Phase 4: Implementation

1. Create Failing Test Case

• Simplest possible reproduction
• Automated test if possible
• MUST have before fixing
• Use TDD skill for proper test

2. Implement Single Fix

• Address the root cause identified
• ONE change at a time
• No "while I'm here" improvements
• No bundled refactoring

3. Verify Fix

• Test passes now?
• No other tests broken?
• Issue actually resolved?

4. If Fix Doesn't Work — 3-Strike Rule

Attempts	Action
< 3	Return to Phase 1, re-analyze with new information
≥ 3	STOP. Show your work. Ask the user.

Anti-rationalization rules:

• "Should work now" → RUN IT. Confidence is not evidence.
• "I already tested earlier" → Code changed since then. Test again.
• "It's a trivial change" → Trivial changes break production. Verify.
• "I'm pretty sure this fixes it" → Pretty sure is not verified. Run the test.

5. After 3+ Failed Fixes: Question Architecture

Pattern indicating architectural problem:

• Each fix reveals new coupling/problem elsewhere
• Fixes require "massive refactoring"
• Each fix creates new symptoms

STOP and question fundamentals:

• Is this pattern fundamentally sound?
• Are we sticking with it through inertia?
• Should we refactor architecture vs. continue fixing symptoms?

Discuss with user before attempting more fixes. Do not attempt a 4th fix without explicit user approval.

Self-Regulation Score (Debug Fix Loops)

When debugging involves multiple fix attempts, track a WTF score to detect runaway fix loops. This complements the 3-Strike Rule above with quantitative drift detection.

Track these signals (default weights, override via ~/.claude-octopus/loop-config.conf):

Event	Score Impact
Revert (git revert, undo, roll back a fix)	+15%
Touching files unrelated to the bug	+20%
A fix that requires changing >3 files	+5%
After the 15th fix attempt	+1% per additional fix
All remaining issues are Low severity	+10%

If WTF score exceeds 20% — STOP immediately, even if under the 3-strike limit. Show the score breakdown and ask the user whether to continue.

Also watch for stuck patterns: If the same error message appears 3+ times across fix attempts, or you see A→B→A→B oscillation (fix X breaks Y, fix Y breaks X), announce the cycle and HALT on second detection.

Report the score with each fix attempt:

Fix attempt 2 | Self-regulation: 15% (1 revert, 0 unrelated files)

Strategy Rotation

After 2 failed fix attempts, stop and reconsider the root cause before trying another fix. If the strategy-rotation hook fires, it means you have been repeating a failing approach. Do not continue down the same path — return to Phase 1 and investigate from a different angle.

Red Flags - STOP and Follow Process

If you catch yourself thinking:

• "Quick fix for now, investigate later"
• "Just try changing X and see"
• "Skip the test, I'll manually verify"
• "It's probably X, let me fix that"
• "I don't fully understand but this might work"
• "One more fix attempt" (when already tried 2+)

ALL of these mean: STOP. Return to Phase 1.

Common Rationalizations

Excuse	Reality
"Issue is simple"	Simple issues have root causes too.
"Emergency, no time"	Systematic is FASTER than thrashing.
"Just try this first"	First fix sets the pattern. Do it right.
"I see the problem"	Seeing symptoms ≠ understanding root cause.
"One more attempt"	3+ failures = architectural problem.

Platform Debugging

If you suspect the issue is with the Claude Code environment itself (e.g., network errors, context limits, tool failures):

• Run /debug: This native command generates a debug bundle to help troubleshoot platform issues.
• Check /debug output: Look for "Context limit", "API error", or "Tool execution failed".

Auto-Freeze on Debug

When debugging a specific module, automatically activate freeze mode to prevent accidental edits outside the investigated area. This is a safety measure that keeps your debugging focused.

How It Works

At the start of Phase 1 (Root Cause Investigation), identify the primary module directory being debugged and activate freeze mode:

# Determine the module directory from the error location or user-specified target
# Example: if debugging src/auth/login.ts, freeze to src/auth/
freeze_dir="$(cd "<module-directory>" 2>/dev/null && pwd)"
echo "${freeze_dir}" > "/tmp/octopus-freeze-${CLAUDE_SESSION_ID:-$$}.txt"

This ensures that during investigation (Phases 1-3), you cannot accidentally modify files outside the module under investigation. When you reach Phase 4 (Implementation), the freeze boundary keeps your fix scoped to the right module.

Auto-freeze activates when:

• The bug is localized to a specific directory (e.g., src/auth/, lib/database/)
• The user specifies a file or module to debug

Auto-freeze does NOT activate when:

• The bug spans multiple modules
• The root cause location is unknown at investigation start
• The user explicitly opts out

After debugging completes, remind the user to run /octo:unfreeze if needed, or remove the state file automatically.

Integration with Claude Octopus

When using octopus workflows for debugging:

Workflow	Debugging Integration
probe	Research error patterns, similar issues
grasp	Define the problem scope clearly
tangle	Implement the fix with TDD
squeeze	Verify fix doesn't introduce vulnerabilities
grapple	Debate architectural alternatives after 3+ failures

Multi-Agent Debugging

For complex bugs, use parallel exploration:

# Phase 1 parallelized
${HOME}/.claude-octopus/plugin/scripts/orchestrate.sh probe "Investigate auth failure from 4 angles"

# Perspectives:
# Agent 1: Error message analysis
# Agent 2: Recent changes review
# Agent 3: Data flow tracing
# Agent 4: Environment comparison

Quick Reference

Phase	Key Activities	Success Criteria
1. Root Cause	Read errors, reproduce, check changes	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

The Bottom Line

Proposing fix → Root cause investigation completed
Otherwise → Not systematic debugging

Systematic approach: 15-30 minutes to fix. Random fixes approach: 2-3 hours of thrashing.

No shortcuts for debugging.