Stop Burning Tokens. Master Claude Code in One Afternoon.
80+ CLAUDE.md templates, 10 hook recipes, 8 custom skill files, subagent patterns, and token optimization workflows. Ship 10x faster without spending 10x more.
12 CLAUDE.md Templates — One For Every Stack
Each template is under 30 lines (the community best practice), battle-tested on real projects, and includes project rules, coding standards, test patterns, and forbidden actions. Drop it in your repo root and Claude Code instantly knows your codebase.
- Component patterns & file structure
- TypeScript strict mode rules
- Testing with Vitest/RTL
- Styling conventions (Tailwind/CSS modules)
- API route patterns
- Pydantic model conventions
- Async/await patterns
- Pytest fixtures & conftest setup
- Virtual env & dependency rules
- Type hint requirements
- Middleware patterns
- Error handling conventions
- ESM vs CommonJS rules
- Auth & validation patterns
- Database query patterns
- Package structure conventions
- Error wrapping patterns
- Interface-first design rules
- Goroutine & channel patterns
- Table-driven test templates
- Ownership & borrowing reminders
- Error handling with thiserror
- Cargo workspace structure
- Unsafe code restrictions
- Documentation test patterns
- MVC conventions enforcement
- Active Record query patterns
- RSpec/FactoryBot setup
- Service object patterns
- Migration safety rules
- Widget composition patterns
- State management rules (Riverpod)
- Platform-specific conventions
- Asset & theming structure
- Integration test setup
- App structure conventions
- Model & manager patterns
- DRF serializer rules
- Celery task patterns
- Template & static file rules
- Composition API patterns
- Pinia store conventions
- Auto-import rules
- SSR/SSG considerations
- E2E test patterns (Playwright)
- Workspace structure (Turborepo/Nx)
- Shared package conventions
- Cross-package import rules
- CI/CD pipeline integration
- Docker compose patterns
- SwiftUI vs UIKit patterns
- MVVM architecture rules
- Core Data conventions
- XCTest setup & patterns
- Accessibility requirements
- Eloquent ORM patterns
- Form request validation
- Queue & job patterns
- Blade template conventions
- PHPUnit test structure
Preview: React/Next.js CLAUDE.md
# Project: [Your App Name] ## Stack React 19 + Next.js 15 (App Router), TypeScript strict, Tailwind CSS Database: Prisma + PostgreSQL | Auth: NextAuth v5 ## Rules - ALWAYS use Server Components by default. Add "use client" only when needed - NEVER use `any` type. Use `unknown` + type narrowing - Components go in /src/components/[Feature]/[Component].tsx - API routes: /src/app/api/[resource]/route.ts - All async functions must have try/catch with proper error types ## Testing - Vitest + React Testing Library for unit tests - Test files: [Component].test.tsx colocated with component - Run: npm test -- --watch ## Forbidden - No default exports except pages/layouts - No console.log in committed code (use logger util) - No direct DOM manipulation - No `git push --force` or `rm -rf`
10 Hook Recipes — Automate Your Workflow
Claude Code hooks run automatically before/after actions. Each recipe is a copy-paste JSON config for your .claude/settings.json. Set it once, never think about it again.
- Blocks rm -rf, git reset --hard
- Prevents force pushes to main
- Warns on .env file access
- Logs all destructive commands
- Customizable blocklist
- Runs Prettier/Black after every edit
- Language-aware (JS, Python, Go)
- Configurable file patterns
- Skips node_modules/vendor
- Zero manual formatting
- Auto-reads key files on session start
- Loads package.json, tsconfig, etc.
- Primes Claude with project context
- Customizable file list
- Reduces "what framework?" questions
- Auto-generates handoff doc on stop
- Lists all files changed
- Captures decisions made
- Creates resume prompt for next session
- Markdown output to /docs
- Runs tests after code changes
- Only tests affected files
- Blocks commit if tests fail
- Shows coverage delta
- Configurable test command
- Enforces conventional commits
- Checks branch naming conventions
- Prevents commits to main directly
- Auto-adds co-author tag
- Validates commit message length
- Warns at 50%, 75%, 90% usage
- Suggests /compact at threshold
- Tracks cost per session
- Daily/weekly cost reports
- Configurable budget limits
- Captures all tool failures
- Writes to .claude/error-log.md
- Includes timestamp & context
- Groups by error type
- Helps debug recurring issues
- Auto-approves read-only commands
- Auto-approves ls, cat, grep, find
- Still prompts for writes/deletes
- Configurable safe command list
- Speeds up exploration 3x
- Desktop notification on completion
- Slack webhook for long tasks
- Email alert for errors
- Sound on task finish
- Configurable per event type
Preview: Security Guard Hook
// .claude/settings.json
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"command": "echo '$TOOL_INPUT' | jq -r '.command' | grep -qE '(rm -rf|git reset --hard|git push.*force|DROP TABLE)' && echo 'BLOCKED: Destructive command detected' && exit 1 || exit 0"
}
]
}
}8 Custom Skill Files — Extend Claude Code
Skills are markdown files in .claude/skills/ that teach Claude Code new capabilities. Drop these files in and invoke them with slash commands.
- Reviews PRs against your team standards
- Checks for security vulnerabilities
- Suggests performance improvements
- Validates test coverage
- Auto-generates from git diff
- Follows your PR template
- Includes test plan section
- Links related issues
- Analyzes error logs & stack traces
- Identifies root cause candidates
- Suggests fix approaches
- Estimates severity & impact
- Write failing test first
- Implement minimal code to pass
- Refactor with confidence
- Tracks red/green/refactor cycle
- Database schema migrations
- API version transitions
- Dependency upgrade paths
- Data transformation scripts
- JSDoc/docstrings from code
- README auto-updater
- API endpoint documentation
- Architecture decision records
- OWASP Top 10 checklist
- Dependency vulnerability scan
- Secret detection patterns
- Input validation review
- Identifies N+1 queries
- Bundle size analysis
- Memory leak detection patterns
- Render performance tips
Bonus: Subagent Patterns & Token Optimization
- Codebase Explorer (gather context without polluting main)
- Parallel File Fixer (independent changes across files)
- Security Reviewer (audit before merge)
- Test Runner (isolated test execution)
- Research Assistant (web search + summarize)
- Code Migrator (bulk updates with validation)
- Slim system prompt technique (41% reduction)
- Strategic /compact timing guide
- Line-range prompting vs. full-file reads
- Model selection: when to use Haiku vs Sonnet vs Opus
- Weekly cost audit checklist
- MCP overhead reduction (67K to under 20K tokens)
- When to start fresh vs. continue
- Context window math (cost per action)
- Signs of session degradation
- Handoff document template
- Git worktree workflow for parallel branches
- Build a feature from scratch
- Debug a production bug
- Refactor a legacy module
- Set up CI/CD from zero
- Write and run a test suite
How We Compare
| Feature | Free Tips | $5 Packs | $795 Courses | Midas Kit |
|---|---|---|---|---|
| CLAUDE.md templates | 3-5 generic | 1-2 | 5-10 | 12 (stack-specific) |
| Hook recipes | 0 | 0 | 3-5 | 10 (copy-paste) |
| Custom skill files | 0 | 0 | 2-3 | 8 ready-to-use |
| Subagent patterns | Blog post | 0 | Theory | 6 with exact prompts |
| Token optimization | Tips | 0 | 1 lecture | Full playbook |
| Time to value | Hours | 30 min | 2 weeks | 15 minutes |
| Price | Free | $5 | $795 | $39 |
FAQ
You're paying $100+/mo for Claude Code.
$39 to actually use it like a pro. Templates, hooks, skills — ready in 15 minutes.
Get the Claude Code Mastery Kit — $39