Introduction
Everything Claude Code (ECC) is a comprehensive configuration collection for Claude Code. It consists of 13 agents, 30+ skills, 30+ commands, rules, and hooks that enable an end-to-end PDCA development cycle.
This cheatsheet serves as a quick reference for every ECC feature.
Slash Commands (/command)
Development Workflow
Command
Description
/plan
Gather requirements, assess risks, and create an implementation plan. Waits for user confirmation before touching code
/tdd
Enforce TDD workflow: write tests first, implement, verify 80%+ coverage
/build-fix
Incrementally fix TypeScript / build errors
/code-review
Security and quality review of uncommitted changes
/refactor-clean
Detect and safely remove dead code (with test verification)
/test-coverage
Analyze test coverage and generate missing tests
/verify
Comprehensive verification of the entire codebase
/checkpoint
Create and verify checkpoints during workflows
E2E Testing
Command
Description
/e2e
Generate and run E2E tests with Playwright
Multi-Model Collaboration
Command
Description
/multi-plan
Collaborative planning with dual models
/multi-execute
Collaborative execution based on a plan
/multi-workflow
Collaborative development with intelligent routing
/multi-frontend
Frontend-focused development workflow
/multi-backend
Backend-focused development workflow
/orchestrate
Sequential agent workflow for complex tasks
Go
Command
Description
/go-test
Go TDD workflow (table-driven tests)
/go-build
Fix Go build errors, vet warnings, and linter issues
/go-review
Review Go code for idiomatic patterns and safety
Python
Command
Description
/python-review
Python review for PEP 8, type hints, and security
Documentation
Command
Description
/update-docs
Sync and update documentation from source
/update-codemaps
Analyze codebase and update architecture documents
Learning System
Command
Description
/learn
Extract and save reusable patterns from sessions
/evolve
Evolve related instincts into skills, commands, or agents
/instinct-status
List all learned instincts with confidence levels
/instinct-export
Export instincts for sharing
/instinct-import
Import instincts
/skill-create
Extract coding patterns from git history and generate SKILL.md
Other
Command
Description
/eval
Evaluation-Driven Development (EDD) workflow management
/sessions
Manage session history (list, load, alias)
/pm2
Analyze project and generate PM2 service commands
/setup-pm
Configure package manager (npm/pnpm/yarn/bun)
Agents (Auto-triggered)
Agents are automatically invoked by Claude Code when the right context arises.
Planning & Architecture
Agent
Trigger
planner
Complex feature requests, refactoring
architect
System design, architectural decisions
Code Quality
Agent
Trigger
code-reviewer
After code changes
python-reviewer
After Python code changes
go-reviewer
After Go code changes
security-reviewer
Changes involving auth, APIs, or sensitive data
database-reviewer
SQL, schema, or DB performance related changes
Testing
Agent
Trigger
tdd-guide
Enforces TDD during new features and bug fixes
e2e-runner
E2E test generation and execution
Fix & Cleanup
Agent
Trigger
build-error-resolver
Build or type errors
go-build-resolver
Go build errors
refactor-cleaner
Dead code detection and removal
Documentation
Agent
Trigger
doc-updater
Documentation updates, CODEMAP generation
Skills (Reference Knowledge)
Skills are the knowledge base that agents and commands draw from.
General
Skill
Content
coding-standards
TypeScript/JS/React/Node.js coding conventions
security-review
Security checklist and patterns
tdd-workflow
TDD workflow (80%+ coverage)
frontend-patterns
React/Next.js, state management, performance
backend-patterns
API design, DB optimization, server-side
eval-harness
Evaluation-Driven Development (EDD) framework
strategic-compact
Optimal timing for context compaction
Python
Skill
Content
python-patterns
PEP 8, type hints, decorators, concurrency
python-testing
pytest, TDD, fixtures, mocking, coverage
Go
Skill
Content
golang-patterns
Idiomatic Go, concurrency patterns
golang-testing
Table-driven tests, benchmarks, fuzzing
Django
Skill
Content
django-patterns
DRF, ORM, caching, signals, middleware
django-security
Auth, CSRF, SQLi, XSS prevention
django-tdd
pytest-django, factory_boy, mocking
Spring Boot
Skill
Content
springboot-patterns
REST API, layered architecture, async processing
springboot-security
Spring Security, auth, rate limiting
springboot-tdd
JUnit 5, Mockito, Testcontainers
jpa-patterns
Entity design, query optimization, transactions
Database
Skill
Content
postgres-patterns
Query optimization, schema design, indexing
clickhouse-io
Analytics workloads, data engineering
Learning
Skill
Content
continuous-learning
Automatic pattern extraction from sessions
continuous-learning-v2
Instinct-based learning system
configure-ecc
ECC interactive installer
Rules (Always Active)
Placed in ~/.claude/rules/, these are enforced at all times.
rules/
├── common/ # Language-agnostic
│ ├── coding-style # Immutability, file organization, error handling
│ ├── git-workflow # Commit messages, PRs, feat/fix/refactor
│ ├── testing # 80%+ coverage, TDD required
│ ├── performance # Model selection, context management
│ ├── patterns # Repository pattern, API response format
│ ├── hooks # PreToolUse/PostToolUse/Stop
│ ├── agents # Agent list and trigger conditions
│ └── security # Secret management, OWASP Top 10
├── python/ # Python-specific
│ ├── coding-style # PEP 8, black/isort/ruff
│ ├── testing # pytest, coverage
│ ├── patterns # Protocol, dataclass
│ ├── hooks # Auto-run black/mypy
│ └── security # bandit, dotenv
└── typescript/ # TypeScript-specific
├── coding-style # Zod, async/await
├── testing # Playwright E2E
├── patterns # Custom hooks, Repository
├── hooks # Auto-run Prettier/tsc
└── security # Environment variables, security-reviewer
Enter fullscreen mode
Exit fullscreen mode
Configuration
File
Scope
Purpose
~/.claude/settings.json
Global
Permissions shared across all projects
<project>/.claude/settings.json
Project (shared)
Team-shared settings
<project>/.claude/settings.local.json
Project (local)
Personal permissions (gitignored)
<project>/CLAUDE.md
Project
Project-specific instructions
Quick Reference
Standard flow for new feature development:
# 1. Plan
/plan
# 2. Implement with TDD
/tdd
# 3. Code review
/code-review
# 4. Final verification
/verify
Enter fullscreen mode
Exit fullscreen mode
When the build breaks:
/build-fix # TypeScript
/go-build # Go
Enter fullscreen mode
Exit fullscreen mode
Refactoring:
/refactor-clean # Remove dead code
/test-coverage # Check coverage
Enter fullscreen mode
Exit fullscreen mode
Learning and improvement:
/learn # Extract patterns
/instinct-status # Check learning status
/evolve # Evolve into skills
Enter fullscreen mode
Exit fullscreen mode
Summary
ECC is a comprehensive framework that dramatically improves development efficiency with Claude Code. Keep this cheatsheet handy for quick reference whenever you need it.
The key takeaways:
Development flow : /plan → /tdd → /code-review → /verify
Auto-triggered agents : Automatically ensure quality at the right moments
Rules : Always active, enforcing coding standards and best practices
Give ECC a try and experience the AI-native development workflow.
For more details, check out the official ECC repository .
Top comments (0)