Auto-Pilot
Updated a month ago

han

TTheBushidoCollective
0.1k
thebushidocollective/han
78
Agent Score

💡 Summary

A CLI tool that automatically runs code quality validations (linting, formatting, type-checking, tests) after Claude Code sessions.

🎯 Target Audience

AI-assisted developers using Claude CodeEngineering leads enforcing code qualityFull-stack developers working across multiple languagesDevOps engineers integrating AI into CI/CD

🤖 AI Roast:It's like a helicopter parent for your AI-generated code, constantly checking its homework before it can go out and play.

Security AnalysisLow Risk

The install method (`curl | bash`) and plugin auto-installation pose supply chain risks, executing untrusted code. The tool also has broad filesystem and network access via plugins for validation. Mitigation: Audit and pin specific plugin versions, run in a sandboxed environment.

Han

GitHub Release

Automatic quality gates for Claude Code. Every conversation ends with validation—linting, formatting, type-checking, and tests run automatically, catching issues before they ship.

Getting Started

Two commands. That's it.

# 1. Install the CLI curl -fsSL https://han.guru/install.sh | bash # 2. Auto-detect and install plugins for your project han plugin install --auto

Next time you use Claude Code, validation hooks run automatically when you finish a conversation.

Alternative Installation

# Homebrew (macOS/Linux) brew install thebushidocollective/tap/han

How It Works

  1. Install - One command installs the CLI and auto-detects plugins for your stack
  2. Code - Claude writes code as usual. No workflow changes needed
  3. Validate - Stop hooks run automatically. Linters, formatters, type checkers, and tests are all verified
  4. Learn - Local metrics track success rates and calibrate confidence. Nothing leaves your machine

Plugin Categories

139 plugins across four categories:

| Category | Description | Examples | |----------|-------------|----------| | Core | Essential infrastructure. Auto-installs han binary, provides metrics and MCP servers | Always required | | Jutsu (Tools) | Validation plugins for your stack | TypeScript, Biome, Pytest, RSpec, ShellCheck | | (Agents) | Specialized AI agents | Code review, debugging, architecture, security | | Hashi (Integrations) | MCP servers for external tools | GitHub, Playwright, Blueprints |

Browse all plugins at han.guru/plugins

Why It Works

  • Smart Caching - Only runs validation when relevant files change. Native Rust hashing keeps it fast
  • Local Metrics - Tracks task success and confidence calibration. All data stays on your machine
  • Zero Config - Binary auto-installs on first session. --auto flag detects your stack automatically
  • Any Stack - TypeScript, Python, Rust, Go, Ruby, Elixir. If there's a linter, there's a plugin

CLI Commands

# Install plugins han plugin install # Interactive mode han plugin install --auto # Auto-detect your stack han plugin install <name> # Install specific plugin # Manage plugins han plugin search <query> # Search marketplace han plugin uninstall <name> # Remove plugin # Run hooks manually han hook run <plugin> <hook> # Run a specific hook han hook explain # Show configured hooks # MCP server han mcp # Start MCP server for natural language hook execution

Documentation

Full documentation at han.guru/docs

Contributing

See CONTRIBUTING.md for how to create new plugins.

License

MIT License - see LICENSE


Built by The Bushido Collective

5-Dim Analysis
Clarity8/10
Novelty7/10
Utility9/10
Completeness7/10
Maintainability8/10
Pros & Cons

Pros

  • Zero-config auto-detection for project stack
  • Runs validation only on changed files for speed
  • Local metrics ensure privacy
  • Extensible plugin ecosystem for many tools

Cons

  • Tightly coupled to Claude Code ecosystem
  • Plugin quality may vary
  • Adds overhead to every coding session
  • Requires trust in auto-installed scripts

Related Skills

dot-skills

A
toolCo-Pilot
82/ 100

“It's a well-organized collection of cheat sheets, but calling them 'skills' might be overselling the automation.”

pytorch

S
toolCode Lib
92/ 100

“It's the Swiss Army knife of deep learning, but good luck figuring out which of the 47 installation methods is the one that won't break your system.”

agno

S
toolCode Lib
90/ 100

“It promises to be the Kubernetes for agents, but let's see if developers have the patience to learn yet another orchestration layer.”

Disclaimer: This content is sourced from GitHub open source projects for display and rating purposes only.

Copyright belongs to the original author TheBushidoCollective.