image

Provider-agnostic code review using AI
Use Claude, Gemini, Codex, OpenCode, Cursor Agent, Kilo, Kiro, Ollama, LM Studio, GitHub Models, MiniMax, or any AI to enforce your coding standards.
Pure Bash core. Works everywhere.

Version License Bash Platforms Homebrew Tests PRs Welcome

Installation • Quick Start • Providers • Commands • Docs

--- ## Example image ## 🎯 Why? You have coding standards. Your team ignores them. Code reviews catch issues too late. **GGA** runs on every commit, validating staged files against your `AGENTS.md`. Like having a senior developer review every line before it hits the repo. ``` ┌─────────────────┐ ┌──────────────┐ ┌─────────────────┐ │ git commit │ ──▶ │ AI Review │ ──▶ │ ✅ Pass/Fail │ │ (staged files) │ │ (any LLM) │ │ (with details) │ └─────────────────┘ └──────────────┘ └─────────────────┘ ``` - 🔌 **Provider agnostic** — Claude, Gemini, Codex, OpenCode, Cursor Agent, Kilo, Kiro, Ollama, LM Studio, GitHub Models, MiniMax - 📦 **Pure Bash core** — no runtime framework; individual providers may require their own CLI or API tooling - 🪝 **Git native** — Standard pre-commit hook - ⚡ **Smart caching** — Skip unchanged files - 🔍 **PR review mode** — Review full PRs, not just last commit - 🪟 **Cross-platform** — macOS, Linux, Windows (Git Bash), WSL --- ## 📦 Installation ### Homebrew (recommended) ```bash brew install gentleman-programming/tap/gga ``` ### Manual ```bash git clone https://github.com/Gentleman-Programming/gentleman-guardian-angel.git cd gentleman-guardian-angel ./install.sh ``` ### Windows (Git Bash, PowerShell, cmd.exe) ```bash git clone https://github.com/Gentleman-Programming/gentleman-guardian-angel.git cd gentleman-guardian-angel bash install.sh echo 'export PATH="$HOME/bin:$PATH"' >> ~/.bashrc && source ~/.bashrc ``` On Windows, the installer also creates `~/bin/gga.bat` so `gga` can be called from `cmd.exe` and PowerShell. Add `%USERPROFILE%\bin` to your Windows user `PATH` for those shells; `.bashrc` only affects Git Bash. > **WSL** is also fully supported — no special configuration needed. ### Oh My Zsh users If you use [Oh My Zsh](https://ohmyz.sh/) with the `git` plugin enabled (the default), the alias `gga` will conflict with this CLI. You'll see: ``` git: 'gui' is not a git command. See 'git --help'. ``` **Fix:** Add this line to your `~/.zshrc` after the Oh My Zsh source line: ```bash unalias gga 2>/dev/null ``` Then run `source ~/.zshrc` or open a new terminal. --- ## 🚀 Quick Start ```bash cd ~/your-project gga init # Create .gga config gga install # Install git hook # Edit .gga to set your PROVIDER # Create AGENTS.md with your coding standards # Done — every commit gets reviewed 🎉 ``` --- ## 🔌 Providers | Provider | Config Value | Installation | |----------|-------------|-------------| | **Claude** | `claude` | [claude.ai/code](https://claude.ai/code) | | **Gemini** | `gemini` | [gemini-cli](https://github.com/google-gemini/gemini-cli) | | **Codex** | `codex` | `npm i -g @openai/codex` | | **OpenCode** | `opencode` | [opencode.ai](https://opencode.ai) | | **Cursor Agent** | `cursor[:model]` | [cursor.com](https://cursor.com) | | **Kilo** | `kilo[:model]` | `npm install -g @kilocode/cli` | | **Kiro** | `kiro` | [kiro.dev/downloads](https://kiro.dev/downloads/) | | **Ollama** | `ollama:` | [ollama.ai](https://ollama.ai) | | **LM Studio** | `lmstudio[:model]` | [lmstudio.ai](https://lmstudio.ai) | | **GitHub Models** | `github:` | [marketplace/models](https://github.com/marketplace/models) | | **MiniMax** | `minimax[:model]` | [platform.minimax.io](https://platform.minimax.io) | > 📖 See [docs/providers.md](docs/providers.md) for detailed examples and setup. --- ## 📋 Commands | Command | Description | |---------|------------| | `gga init` | Create sample `.gga` config | | `gga install` | Install pre-commit hook | | `gga install --commit-msg` | Install commit-msg hook | | `gga uninstall` | Remove hooks | | `gga run` | Review staged files | | `gga run --ci` | Review last commit (CI/CD) | | `gga run --pr-mode` | Review full PR changes | | `gga run --no-cache` | Review ignoring cache | | `gga config` | Show configuration | | `gga cache status` | Show cache info | | `gga version` | Show version | > 📖 See [docs/commands.md](docs/commands.md) for detailed usage. --- ## 📚 Documentation | Topic | Description | |-------|------------| | [Configuration](docs/configuration.md) | `.gga` config file, options, hierarchy, env overrides | | [Rules File](docs/rules-file.md) | Writing effective `AGENTS.md`, best practices, skill-based approach | | [Providers](docs/providers.md) | Detailed setup for each AI provider | | [Commands](docs/commands.md) | Full command reference with examples | | [Caching](docs/caching.md) | How smart caching works, invalidation, commands | | [Integrations](docs/integrations.md) | Husky, pre-commit, Lefthook, VS Code, CI/CD | | [Examples](docs/examples.md) | Real-world walkthrough, project configs | | [Troubleshooting](docs/troubleshooting.md) | Common issues and fixes | | [Changelog](docs/changelog.md) | Version history | | [Contributing](CONTRIBUTING.md) | How to contribute (issue-first workflow) | --- ## 📄 License MIT © 2024

Built with 🧉 by developers who got tired of repeating the same code review comments