Error Resolver Skill for Claude

Error Resolver

Legacy reference archive for first-principle error diagnosis across languages and frameworks.

Routing Status

This directory is retained because it contains substantial analysis, pattern, and replay material. It is not an active code-quality route owner. Route user-facing bug, stack-trace, and root-cause tasks to systematic-debugging.

Core Philosophy

The 5-step Error Resolution Process:

1. CLASSIFY  ->  2. PARSE  ->  3. MATCH  ->  4. ANALYZE  ->  5. RESOLVE
     |              |             |             |              |
  What type?    Extract key    Known       Root cause      Fix +
               information    pattern?     analysis       Prevent

Quick Start

When you encounter an error:

• Paste the full error (including stack trace if available)
• Provide context (what were you trying to do?)
• Share relevant code (the file/function involved)

Error Classification Framework

Primary Categories

| Category | Indicators | Common Causes |

|----------|------------|---------------|

| Syntax | Parse error, Unexpected token | Typos, missing brackets, invalid syntax |

| Type | TypeError, type mismatch | Wrong data type, null/undefined access |

| Reference | ReferenceError, NameError | Undefined variable, scope issues |

| Runtime | RuntimeError, Exception | Logic errors, invalid operations |

| Network | ECONNREFUSED, timeout, 4xx/5xx | Connection issues, wrong URL, server down |

| Permission | EACCES, PermissionError | File/directory access, sudo needed |

| Dependency | ModuleNotFound, Cannot find module | Missing package, version mismatch |

| Configuration | Config error, env missing | Wrong settings, missing env vars |

| Database | Connection refused, query error | DB down, wrong credentials, bad query |

| Memory | OOM, heap out of memory | Memory leak, large data processing |

Secondary Attributes

• Severity: Fatal / Error / Warning / Info
• Scope: Build-time / Runtime / Test-time
• Origin: User code / Framework / Third-party / System

Analysis Workflow

Step 1: Classify

Identify the error category by examining:

• Error name/code (e.g., ENOENT, TypeError)
• Error message keywords
• Where it occurred (compile, runtime, test)

Step 2: Parse

Extract key information:

- Error code: [specific code if any]
- File path: [where the error originated]
- Line number: [exact line if available]
- Function/method: [context of the error]
- Variable/value: [what was involved]
- Stack trace depth: [how deep is the call stack]

Step 3: Match Patterns

Check against known error patterns:

• See patterns/ directory for language-specific patterns
• Match error signatures to known solutions
• Check replay history for previous solutions

Step 4: Root Cause Analysis

Apply the 5 Whys technique:

Error: Cannot read property 'name' of undefined
  Why 1? -> user object is undefined
  Why 2? -> API call returned null
  Why 3? -> User ID doesn't exist in database
  Why 4? -> ID was from stale cache
  Why 5? -> Cache invalidation not implemented

Root Cause: Missing cache invalidation logic

Step 5: Resolve

Generate actionable solution:

• Immediate fix - Get it working now
• Proper fix - The right way to solve it
• Prevention - How to avoid in the future

Output Format

When resolving an error, provide:

## Error Diagnosis

**Classification**: [Category] / [Severity] / [Scope]

**Error Signature**:
- Code: [error code]
- Type: [error type]
- Location: [file:line]

## Root Cause

[Explanation of why this error occurred]

**Contributing Factors**:
1. [Factor 1]
2. [Factor 2]

## Solution

### Immediate Fix
[Quick steps to resolve]

### Code Change
[Specific code to add/modify]

### Verification
[How to verify the fix works]

## Prevention

[How to prevent this error in the future]

## Replay Tag

[Unique identifier for this solution - for future reference]

Replay System

The replay system records successful solutions for future reference.

Recording a Solution

After resolving an error, record it:

# Create solution record in project
mkdir -p .claude/error-solutions

# Solution file format: [error-type]-[hash].yaml

Solution Record Format

# .claude/error-solutions/[error-signature].yaml
id: "nodejs-module-not-found-express"
created: "2024-01-15T10:30:00Z"
updated: "2024-01-20T14:22:00Z"

error:
  type: "dependency"
  category: "ModuleNotFound"
  language: "nodejs"
  pattern: "Cannot find module 'express'"
  context: "npm project, missing dependency"

diagnosis:
  root_cause: "Package not installed or node_modules corrupted"
  factors:
    - "Missing npm install after git clone"
    - "Corrupted node_modules directory"
    - "Package not in package.json"

solution:
  immediate:
    - "Run: npm install express"
  proper:
    - "Check package.json has express listed"
    - "Run: rm -rf node_modules && npm install"
  code_change: null

verification:
  - "Run the application again"
  - "Check express is in node_modules"

prevention:
  - "Add npm install to project setup docs"
  - "Use npm ci in CI/CD pipelines"

metadata:
  occurrences: 5
  last_resolved: "2024-01-20T14:22:00Z"
  success_rate: 1.0
  tags: ["nodejs", "npm", "dependency"]

Replay Lookup

When encountering an error:

• Generate error signature from the error message
• Search .claude/error-solutions/ for matching patterns
• If found, apply the recorded solution
• If new, proceed with full analysis and record the solution

Error Signature Generation

signature = hash(
  error_type +
  error_code +
  normalized_message +  # remove specific values
  language +
  framework
)

Example transformations:

• Cannot find module 'express' -> Cannot find module '{module}'
• TypeError: Cannot read property 'name' of undefined -> TypeError: Cannot read property '{prop}' of undefined

Debug Commands

Useful commands during debugging:

Node.js

# Verbose error output
NODE_DEBUG=* node app.js

# Memory debugging
node --inspect app.js

# Check installed packages
npm ls [package-name]

# Verify package.json
npm ls --depth=0

Python

# Debug mode
python -m pdb script.py

# Check installed packages
pip show [package-name]
pip list

General

# Check file permissions
ls -la [file]

# Check port usage
lsof -i :[port]
netstat -an | grep [port]

# Check environment variables
env | grep [VAR_NAME]
printenv [VAR_NAME]

# Check disk space
df -h

# Check memory
free -m  # Linux
vm_stat  # macOS

Common Debugging Patterns

Pattern 1: Binary Search

When the error location is unclear:

• Comment out half the code
• If error persists, it's in the remaining half
• Repeat until you find the exact line

Pattern 2: Minimal Reproduction

Create the smallest code that reproduces the error:

• Start with empty file
• Add code piece by piece
• Stop when error appears
• That's your minimal repro case

Pattern 3: Rubber Duck Debugging

Explain the problem out loud (or to Claude):

• What should happen?
• What actually happens?
• What changed recently?
• What assumptions am I making?

Pattern 4: Git Bisect

Find which commit introduced the bug:

git bisect start
git bisect bad  # current commit is bad
git bisect good [last-known-good-commit]
# Git will checkout commits for you to test
git bisect good/bad  # mark each as good or bad
git bisect reset  # when done

Reference Files

• patterns/ - Language-specific error patterns
• nodejs.md - Node.js common errors
• python.md - Python common errors
• react.md - React/Next.js errors
• database.md - Database errors
• docker.md - Docker/container errors
• git.md - Git errors
• network.md - Network/API errors

• analysis/ - Analysis methodologies
• stack-trace.md - Stack trace parsing guide
• root-cause.md - Root cause analysis techniques

• replay/ - Replay system
• solution-template.yaml - Template for recording solutions