Frequently Asked Questions about aicommit

Q: What is aicommit?

A: Aicommit is a command-line tool that uses Large Language Models (LLMs) to generate concise and descriptive Git commit messages automatically.

Q: How does aicommit work?

A: Aicommit analyzes your code changes using LLMs and generates a commit message that accurately reflects the modifications you’ve made.

Q: What LLM providers are supported by aicommit?

A: Aicommit supports OpenRouter, Simple Free OpenRouter, Ollama, and OpenAI-compatible API endpoints.

Q: What is Simple Free Mode in aicommit?

A: Simple Free Mode allows you to use OpenRouter’s free models without needing to manually select them. Aicommit automatically selects the best available free model and handles failover and performance tracking.

Q: How do I install aicommit?

A: You can install aicommit using npm with the command npm install -g @suenot/aicommit or using cargo with cargo install aicommit.

Q: How do I generate a commit message with aicommit?

A: After installing aicommit, stage your changes with git add . and then run aicommit to generate a commit message. Alternatively, you can use aicommit --add to automatically stage and commit your changes.

Q: Can I customize the generated commit messages?

A: Yes, you can customize the behavior of aicommit by editing the configuration file located at ~/.aicommit.json. You can modify settings like the active LLM provider, maximum tokens, and temperature.

Q: Is there a VS Code extension for aicommit?

A: Yes, there is a VS Code extension available for seamless integration with the editor. See the VS Code Extension README for more details.

Q: What is the watch mode in aicommit?

A: The watch mode allows you to automatically commit changes when files are modified. Use aicommit --watch to continuously monitor files and commit changes as they occur.

Q: How do I manage model jails and blacklists?

A: Aicommit provides commands for managing model jails and blacklists. Use aicommit --jail-status to show the current status of all models. Use aicommit --unjail="model_id" to release a specific model, and aicommit --unjail-all to release all models.

Q: How can I configure aicommit to use LM Studio as an OpenAI-compatible API endpoint?

A: Start LM Studio, load a model, and start the server. Then, configure aicommit by selecting ‘OpenAI Compatible’ and setting the API Key to lm-studio, API URL to http://localhost:1234/v1/chat/completions (or the URL shown in LM Studio), and Model to lm-studio-model (or any descriptive name).

Q: What does ‘Tokens: X↑ Y↓’ mean in the output?

A: In the output, ‘Tokens: X↑ Y↓’ indicates the number of tokens used, where X↑ is the number of input tokens and Y↓ is the number of output tokens.