git-acm configuration guide

This guide explains how to configure and set up the git-acm tool for generating AI-powered commit messages.

Prerequisites

Before configuring git-acm, ensure you have:

• Git initialized in your project directory
• At least one of the supported AI API keys
• A .env file in your project root (recommended) or environment variables set

Available APIs

git-acm supports multiple models:

• claude sonnet 4
• claude opus 4
• o1
• gpt-4o
• gpt-4.1
• o1-pro
• gpt-4.1-mini
• gpt-4.1-nano
• claude-3.7-sonnet
• claude-3.5-sonnet
• claude-3.5-haiku
• gemini-2.5-pro
• gemini-2.0-flash
• gemini-2.0-flash-lite
• deepseek-chat
• deepseek-reasoner
• llama-4-scout
• llama-4-maverick
• llama-3.3-70b-instruct
• llama-3.1-8b-instruct
• llama-3.1-70b-instruct
• llama-3.2-3b-instruct

Setting Up API Keys

1. Get Your API Keys

   • gemini: Visit google AI Studio
   • openai: Visit openai Platform
   • anthropic: Visit anthropic console
   • llama: use ollama
   • deepseek: visit deepseek

2. Configure Environment Variables

   Add the following to your project’s .env file or export them in your terminal:

   .env

   # Gemini Configuration
   GEMINI_API_KEY="your-gemini-api-key"
   
   # Anthropic Configuration
   ANTHROPIC_API_KEY="your-anthropic-api-key"
   
   # OpenAI Configuration
   OPENAI_API_KEY="your-openai-api-key"
   
   # Llama Configuration (using Ollama)
   LLAMA_API_URL="http://localhost:11434/api/generate"
   LLAMA_MODEL_NAME="llama3.2:1b"
   
   # for deepseek api
   DEEPSEEEK_API_KEY=""

Selecting an API Provider

after setting up the api keys, select the model you want to use:

git-acm use <model_name>

Replace <model_name> with one of:

- claude sonnet 4
- claude opus 4
- o1
- gpt-4o
- gpt-4.1
- o1-pro
- gpt-4.1-mini
- gpt-4.1-nano
- claude-3.7-sonnet
- claude-3.5-sonnet
- claude-3.5-haiku
- gemini-2.5-pro
- gemini-2.0-flash
- gemini-2.0-flash-lite
- deepseek-chat
- deepseek-reasoner
- llama-4-scout
- llama-4-maverick
- llama-3.3-70b-instruct
- llama-3.1-8b-instruct
- llama-3.1-70b-instruct
- llama-3.2-3b-instruct

Auto-commit Configuration

git-acm includes an auto-commit feature that can be enabled or disabled:

# Enable auto-commit
git-acm autocommit enable

# Disable auto-commit
git-acm autocommit disable

When enabled, git-acm will automatically commit changes with the generated message.

Usage After Configuration

Once configured, simply run:

git-acm

This will generate a commit message using your selected API provider.

Troubleshooting

If you encounter issues:

1. Verify your API keys are correctly set in the .env file
2. Ensure Git is initialized in your project directory
3. Check that you’ve selected an modelusing git-acm use
4. For Llama, ensure Ollama is running locally
5. Verify your internet connection for cloud API providers

Support

For bug reports or issues, contact me:

• twitter: @shivamhwp