Getting Started with giv
Your first steps to AI-powered Git workflows
Get giv installed and configured in under 5 minutes. Transform your Git history into professional documentation with zero friction.
Quick Start Guide
From zero to AI-powered commits in 3 steps
Install giv
# One-line install script (recommended) curl -fsSL https://raw.githubusercontent.com/fwdslsh/giv/main/install.sh | sh # Or install via pip pip install giv # Or use Docker docker pull fwdslsh/giv:latest
The install script automatically detects your platform and downloads the appropriate binary. No dependencies required.
Configure AI Provider
# Set up OpenAI (most common) export OPENAI_API_KEY="your-api-key" export GIV_API_MODEL="gpt-4" # Or configure via giv giv config set api.url "https://api.openai.com/v1/chat/completions" giv config set api.model "gpt-4"
Support for OpenAI, Anthropic, Ollama, and custom endpoints. Environment variables are the quickest way to get started.
Generate Your First Commit
# Initialize giv in your project giv init # Make some changes and stage them git add . # Generate AI-powered commit message giv message
giv analyzes your staged changes and generates a professional commit message following best practices.
AI Provider Configuration
Choose your preferred AI service
OpenAI Configuration
# Environment variables (recommended) export OPENAI_API_KEY="sk-your-api-key-here" export GIV_API_MODEL="gpt-4" # Or use giv config giv config set api.url "https://api.openai.com/v1/chat/completions" giv config set api.model "gpt-4" giv config set api.key "sk-your-api-key-here"
Recommended models: gpt-4, gpt-4-turbo, gpt-3.5-turbo
Get your API key from OpenAI Platform
Anthropic Claude Configuration
# Environment variables export ANTHROPIC_API_KEY="your-anthropic-key" export GIV_API_MODEL="claude-3-5-sonnet-20241022" # Or use giv config giv config set api.url "https://api.anthropic.com/v1/messages" giv config set api.model "claude-3-5-sonnet-20241022" giv config set api.key "your-anthropic-key"
Recommended models: claude-3-5-sonnet-20241022, claude-3-opus-20240229
Get your API key from Anthropic Console
Ollama (Local) Configuration
# Install Ollama first curl -fsSL https://ollama.ai/install.sh | sh # Pull a model ollama pull llama3.2 # Configure giv giv config set api.url "http://localhost:11434/v1/chat/completions" giv config set api.model "llama3.2" giv config set api.key "not-needed"
Recommended models: llama3.2, codellama, mistral
Completely private and offline after initial model download
Custom API Configuration
# Any OpenAI-compatible endpoint giv config set api.url "https://your-custom-endpoint.com/v1/chat/completions" giv config set api.model "your-model-name" giv config set api.key "your-api-key" # Optional: custom headers giv config set api.headers.Authorization "Bearer your-token" giv config set api.headers.Custom-Header "custom-value"
Works with any service that implements the OpenAI chat completions API
Essential Commands
Core giv commands you'll use every day
Workflow Integration
Integrate giv into your development process
Git Hooks
# Add to .git/hooks/prepare-commit-msg
#!/bin/sh
if [ -z "$2" ]; then
giv message > "$1"
fi
Automatically generate commit messages during git commit
CI/CD Integration
# GitHub Actions example
- name: Generate Release Notes
run: |
giv release-notes ${{ github.event.release.tag_name }} \
--output release-notes.md
Automate documentation generation in your CI/CD pipeline
Custom Templates
# Initialize project templates giv init # Edit custom prompts nano .giv/templates/commit_message_prompt.md
Customize AI prompts for your project's specific needs
⚠️ Common Issues & Solutions
GLIBC Version Error (Linux)
# Use pip installation instead pip install giv # Or check compatibility ldd --version
Binary requires GLIBC 2.31+ (Ubuntu 20.04+)
API Rate Limits
# Add delays between requests giv config set api.delay 1000 # Or use a different model giv config set api.model "gpt-3.5-turbo"
Configure rate limiting for your API tier