Claude Code Plugins

Community-maintained marketplace

Feedback

Analyze codebase for patterns, issues, and improvements. Enforces analysis standards while incorporating user-specific requirements.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name code-analyze
description Analyze codebase for patterns, issues, and improvements. Enforces analysis standards while incorporating user-specific requirements.

Code Analyzer Skill

What This Skill Does

Performs standardized code analysis with user context:

  • Analyzes codebase structure and patterns
  • Identifies technical debt and issues
  • Suggests improvements and refactoring
  • Enforces coding standards
  • Generates actionable reports

When Claude Should Use This

Use this skill when the user:

  • Says "analyze the code" or "review the codebase"
  • Wants to find bugs or issues
  • Needs architecture review
  • Mentions code quality or technical debt
  • Requests security or performance analysis

Analysis Workflow

1. Gather User Context

Ask the user:
1. Focus areas? (security, performance, architecture, all)
2. Specific concerns?
3. Depth of analysis? (quick scan, standard, deep dive)
4. Output format? (report, tasks, PR comments)

2. Standardized Analysis Process

Phase 1: Structure Analysis

# Map codebase structure
find . -type f -name "*.ts" -o -name "*.js" | head -20

# Count lines of code
cloc . --exclude-dir=node_modules,dist,out

# Identify entry points
grep -r "export.*function\|export.*class" --include="*.ts" | head -10

Phase 2: Pattern Detection

# Find potential issues
grep -r "TODO\|FIXME\|HACK\|XXX" --include="*.ts"

# Identify deprecated patterns
grep -r "deprecated\|@deprecated" --include="*.ts"

# Find console.logs (potential debug code)
grep -r "console\." --include="*.ts" --exclude-dir=node_modules

Phase 3: Dependency Analysis

# Check for outdated packages
npm outdated

# Audit for vulnerabilities
npm audit

# Analyze bundle size
npm ls --depth=0

3. Automated Report Generation

Generate analysis/ANALYSIS_[DATE].md:

# Code Analysis Report

## Summary
- **Date**: YYYY-MM-DD
- **Scope**: [User specified focus]
- **Health Score**: X/100

## Critical Issues
1. [Issue]: [Description]
   - File: [path:line]
   - Severity: High/Medium/Low
   - Fix: [Recommendation]

## Architecture Review
- **Patterns Found**: [List]
- **Anti-patterns**: [List]
- **Suggestions**: [List]

## Technical Debt
- **Total**: X hours estimated
- **Priority Items**: [List]

## Action Items
- [ ] Fix critical security issues
- [ ] Refactor deprecated patterns
- [ ] Update dependencies

4. Integration with Git Workflow

For Issues Found:

# Create issue branch
git checkout -b fix/[issue-name]

# For refactoring needs
git checkout -b refactor/[component-name]

# For security fixes
git checkout -b security/[vulnerability-name]

5. Enforcement Rules

Must Fix (Blocking):

  • Security vulnerabilities (High/Critical)
  • Breaking changes in dependencies
  • Memory leaks
  • Exposed secrets/credentials

Should Fix (Warning):

  • Deprecated API usage
  • Performance bottlenecks
  • Code duplication > 50 lines
  • Circular dependencies

Consider Fixing (Info):

  • TODO/FIXME comments
  • Missing documentation
  • Complex functions (cyclomatic > 10)
  • Long files (> 500 lines)

Analysis Types

1. Security Analysis

# Check for secrets
grep -r "api[_-]key\|password\|secret\|token" --include="*.ts"

# Review authentication
grep -r "authenticate\|authorize\|jwt\|session" --include="*.ts"

# Check HTTPS usage
grep -r "http://" --include="*.ts"

2. Performance Analysis

# Find potential bottlenecks
grep -r "for.*for\|while.*while" --include="*.ts"

# Check for synchronous operations
grep -r "readFileSync\|execSync" --include="*.ts"

# Identify heavy operations
grep -r "sort\|filter.*map\|reduce" --include="*.ts"

3. Architecture Analysis

  • Component coupling
  • Layer violations
  • Dependency cycles
  • Interface segregation
  • Single responsibility

Output Actions

1. Create Fix Branches

For each critical issue:

git checkout -b fix/[issue-id]
echo "Fix plan" > fixes/[issue-id].md

2. Generate Tasks

Create tasks/FIXES_[DATE].md:

## Critical Fixes Required

### HIGH Priority
- [ ] Fix SQL injection in [file:line]
- [ ] Remove hardcoded credentials in [file:line]

### MEDIUM Priority
- [ ] Refactor [component] to reduce complexity
- [ ] Update deprecated [package] usage

### LOW Priority
- [ ] Add missing documentation
- [ ] Clean up TODO comments

3. PR Templates

Generate .github/PULL_REQUEST_TEMPLATE/fix.md:

## Fix for: [Issue ID]

### What was wrong?
[Description]

### How was it fixed?
[Approach]

### Testing
- [ ] Unit tests pass
- [ ] Integration tests pass
- [ ] Security scan clean

Integration with User Intent

The analyzer combines:

  1. User's specific concerns (from prompt)
  2. Standardized checks (always run)
  3. Codebase context (patterns, history)
  4. Best practices (industry standards)

Example:

  • User: "Check our API for security issues"
  • Analyzer: Runs standard checks + deep API security scan

Protection Against Breaking Changes

Allowed Fixes:

  • Security patches that don't change interfaces
  • Performance improvements with same behavior
  • Bug fixes with tests
  • Documentation updates

Requires Approval:

  • API changes
  • Database schema changes
  • Configuration changes
  • Dependency major version updates

Never Auto-Fix:

  • Core business logic
  • Authentication/Authorization
  • Payment processing
  • Data migrations

Success Metrics

Analysis succeeds when:

  • All critical issues identified
  • Report generated with actionable items
  • Fix branches created for issues
  • Team understands findings
  • Clear path to resolution