██████╗██╗ █████╗ ██╗ ██╗██████╗ ███████╗ ██████╗ ██████╗ ██████╗ ███████╗
██╔════╝██║ ██╔══██╗██║ ██║██╔══██╗██╔════╝ ██╔════╝██╔═══██╗██╔══██╗██╔════╝
██║ ██║ ███████║██║ ██║██║ ██║█████╗ ██║ ██║ ██║██║ ██║█████╗
██║ ██║ ██╔══██║██║ ██║██║ ██║██╔══╝ ██║ ██║ ██║██║ ██║██╔══╝
╚██████╗███████╗██║ ██║╚██████╔╝██████╔╝███████╗ ╚██████╗╚██████╔╝██████╔╝███████╗
╚═════╝╚══════╝╚═╝ ╚═╝ ╚═════╝ ╚═════╝ ╚══════╝ ╚═════╝ ╚═════╝ ╚═════╝ ╚══════╝
██████╗ ██╗ ██╗ ██╗ ██████╗ ██╗███╗ ██╗███████╗
██╔══██╗██║ ██║ ██║██╔════╝ ██║████╗ ██║██╔════╝
██████╔╝██║ ██║ ██║██║ ███╗██║██╔██╗ ██║███████╗
██╔═══╝ ██║ ██║ ██║██║ ██║██║██║╚██╗██║╚════██║
██║ ███████╗╚██████╔╝╚██████╔╝██║██║ ╚████║███████║
╚═╝ ╚══════╝ ╚═════╝ ╚═════╝ ╚═╝╚═╝ ╚═══╝╚══════╝
Productivity commands for Claude Code CLI that save 2-3 hours per week on repetitive tasks.
🤦 Ask Claude to fix a bug → Get 15 test files
😤 Request a simple refactor → Receive a dissertation on clean code
🙄 "Please add a button" → Complete UI framework rewrite
😭 Every conversation → "Act like a google engineer who doesn't overengineer"
CCPlugins is a curated set of commands that extend Claude Code Cli with common development workflows. These commands handle the dumb work, since the agent (opus 4 / sonnet 4 or kimi k2) already knows your codebase, they just tell it what to do with that knowledge.
- Installation - Get started in 30 seconds
- Commands - See all available commands
- How It Works - Understanding the magic
- Technical Notes - Why conversational design matters
- Contributing - Help make it better
Mac/Linux:
curl -sSL https://raw.githubusercontent.com/brennercruvinel/CCPlugins/main/install.sh | bashWindows/Cross-platform:
python install.pygit clone https://github.com/brennercruvinel/CCPlugins.git
cd CCPlugins
python install.py# Mac/Linux
./uninstall.sh
# Windows/Cross-platform
python uninstall.pyPre-configured commands that make Claude Code work like the senior engineer you keep asking for.
Remove debug artifacts and clean up after development sessions
/cleanproject
Analyze changes and create conventional commit messages
/commit
Auto-detect and run the project's code formatter
/format
Run tests and automatically fix simple failures
/test
Comprehensive code review for bugs, security, and performance
/review
Remove obvious comments while preserving valuable documentation
/remove-comments
Remove TypeScript any types and suggest proper types
/cleanup-types
Fix broken imports after moving or renaming files
/fix-imports
Find all TODO, FIXME , and HACK comments in your codebase
/find-todos
Begin a documented coding session with goals tracking
/session-start
Summarize accomplishments and prepare handoff notes
/session-end
Rollback last operation with automatic backup restore
/undo
Cache project context to speed up subsequent commands
/context-cache
src/
├── UserService.js
├── UserService.test.js
├── UserService_backup.js # Old version
├── debug.log # Debug output
├── test_temp.js # Temporary test
└── notes.txt # Dev notes
src/
├── UserService.js # Clean production code
└── UserService.test.js # Actual tests preserved
CCPlugins are markdown files that provide intelligent instructions to Claude Code. When you type a command:
- Claude reads the command definition from
~/.claude/commands/ - Analyzes your project context
- Executes the appropriate actions
- Provides clear feedback
Key Principles:
- Commands are conversational instructions, not rigid scripts
- Claude interprets and adapts to your specific project context
- Multiple tools can be used in parallel for efficiency
- Works with any language or framework through contextual analysis
Limitations:
- Commands rely on Claude's contextual interpretation
- May need guidance in unconventional project structures
- Subject to model usage limits (Opus/Sonnet)
Commands are written in first person ("I'll help you...") rather than imperative ("Do this..."). This design choice transforms Claude Code from a command executor into a collaborative assistant, creating a more conversational and helpful interaction.
✅ Good: "I'll analyze your code and fix broken imports..."
❌ Avoid: "Analyze code and fix broken imports..."This approach:
- Makes Claude feel like a
"partner", not a tool (and for some reason, it actually works better this way) - Sets
clear expectationsabout what will happen - Creates a more natural,
human-like interaction Reduces the intimidationfactor for new users
Note: Basic testing with Kimi K2 shows excellent compatibility, but more validation and testing is needed to ensure full accuracy across all commands.
Custom commands appear with a (user) tag in Claude Code CLI to distinguish them from built-in commands. This is normal and indicates your commands are properly installed.
/commit
Smart Git Commit (user) ← Your custom command
/help
Show help ← Built-in command
| Task | Manual Time | With Commands | Time Saved |
|---|---|---|---|
| Git commits | 5-10 min | 30 sec | ~9 min |
| Code cleanup | 20-30 min | 1 min | ~25 min |
| Test fixes | 15-20 min | 2-5 min | ~15 min |
| Code review | 20 min | 2 min | ~18 min |
| Rollback mistakes | 10-15 min | 30 sec | ~12 min |
Average: 2-3 hours saved per week
- Claude Code CLI
- Python 3.6+ (for installer)
- Git (for version control commands)
Create your own commands by adding markdown files to ~/.claude/commands/:
# My Custom Command
I'll help you with your specific workflow.
[Your instructions here]Commands support arguments via $ARGUMENTS:
/mycommand some-file.js
# $ARGUMENTS will contain "some-file.js"Commands work seamlessly in automated workflows:
claude /test && claude /commitWe welcome contributions that help developers save time. See CONTRIBUTING.md for guidelines.
MIT License - see LICENSE file for details.
Built by a developer tired of typing please act like a senior engineer in every conversation.