Automated git commit messages using GPT models via gptcommit for VS Code.

• Install the extension from the VS Code Marketplace
• Install the gptcommit

Note: Do NOT install gptcommit hook via gptcommit install under the root of your git repo.

Run via Ctrl+Shift+P or Cmd+Shift+P:

• GPTCommit: Generate Git Commit Message Generate the commit message

• GPTCommit: Setup OpenAI API Key Setup the OpenAI API Key. You can get the API key from OpenAI

• GPTCommit: Use a different OpenAI model Use a different OpenAI model. For a list of public OpenAI models, checkout the OpenAI docs. Default is now gpt-3.5-turbo.

• GPTCommit: Set output language Set the output language. Default is en.

• GPTCommit: Show per-file summary Enable "show per-file summary"? It's disabled by default.

• GPTCommit: Disable conventional commit Disable "conventional commit"? It's enabled by default.

• GPTCommit: Open gptcommit configuration file Open the local gptcommit configuration file (~/.git/gptcommit.toml)

• ExpressMode: If true, generated message will be filled into the Source Control commit message input box directly, instead of opening a new editor.
• ExpressModeContent: Content of the message to fill in the express mode.
  • Note that to show per-file summary, you need to enable "show per-file summary" by running the GPTCommit: Show per-file summary command.
• GptcommitPath: Path to the gptcommit executable.
• OnFiles: Diff of files to use for generating the commit message.
  • staged: Use staged files
  • unstaged: Use unstaged files
  • tryStagedThenUnstaged: Try staged files first, then try unstaged files if no staged files are found

Note that now all the configuration via the extension is saved in the .git/gptcommit.toml file. If you have to change advanced configuration, you can edit the .git/gptcommit.toml file directly, but make sure you know what you are doing. You can also use the GPTCommit: Open gptcommit configuration file command to open the configuration file.

If you want to use the configuration globally, you can copy the .git/gptcommit.toml file to ~/.config/gptcommit/config.toml, or just the sections of the configuration you want to be used globally.

Also refer to the gptcommit documentation for more information.