AICommit2

A Reactive CLI that generates git commit messages with Ollama, ChatGPT, Gemini, Claude, Mistral and other AI

aicommit2 is a reactive CLI tool that automatically generates Git commit messages using various AI models. It supports simultaneous requests to multiple AI providers, allowing users to select the most suitable commit message. The core functionalities and architecture of this project are inspired by AICommits.

• Multi-AI Support: Integrates with OpenAI, Anthropic Claude, Google Gemini, Mistral AI, Cohere, Groq and more.
• Local Model Support: Use local AI models via Ollama.
• Reactive CLI: Enables simultaneous requests to multiple AIs and selection of the best commit message.
• Git Hook Integration: Can be used as a prepare-commit-msg hook.
• Custom Prompt: Supports user-defined system prompt templates.

• OpenAI
• Anthropic Claude
• Gemini
• Mistral AI (including Codestral)
• Cohere
• Groq
• Perplexity
• DeepSeek
• Huggingface (Unofficial)

• Ollama

The minimum supported version of Node.js is the v18. Check your Node.js version with node --version.

1. Install aicommit2:

npm install -g aicommit2

2. Set up API keys (at least ONE key must be set):

aicommit2 config set OPENAI.key=<your key>
aicommit2 config set ANTHROPIC.key=<your key>
# ... (similar commands for other providers)

3. Run aicommit2 with your staged files in git repository:

git add <files...>
aicommit2

👉 Tip: Use the aic2 alias if aicommit2 is too long for you.

You can also use your model for free with Ollama and it is available to use both Ollama and remote providers simultaneously.

1. Install Ollama from https://ollama.com

2. Start it with your model

ollama run llama3.2 # model you want use. ex) codellama, deepseek-coder

3. Set the model and host

aicommit2 config set OLLAMA.model=<your model>

If you want to use Ollama, you must set OLLAMA.model.

4. Run aicommit2 with your staged in git repository

git add <files...>
aicommit2

👉 Tip: Ollama can run LLMs in parallel from v0.1.33. Please see this section.

This CLI tool runs git diff to grab all your latest code changes, sends them to configured AI, then returns the AI generated commit message.

If the diff becomes too large, AI will not function properly. If you encounter an error saying the message is too long or it's not a valid commit message, try reducing the commit unit.

You can call aicommit2 directly to generate a commit message for your staged changes:

git add <files...>
aicommit2

aicommit2 passes down unknown flags to git commit, so you can pass in commit flags.

For example, you can stage all changes in tracked files with as you commit:

aicommit2 --all # or -a

• --locale or -l: Locale to use for the generated commit messages (default: en)
• --all or -a: Automatically stage changes in tracked files for the commit (default: false)
• --type or -t: Git commit message format (default: conventional). It supports conventional and gitmoji
• --confirm or -y: Skip confirmation when committing after message generation (default: false)
• --clipboard or -c: Copy the selected message to the clipboard (default: false).
  • If you give this option, aicommit2 will not commit.
• --generate or -g: Number of messages to generate (default: 1)
  • Warning: This uses more tokens, meaning it costs more.
• --exclude or -x: Files to exclude from AI analysis

Example:

aicommit2 --locale "jp" --all --type "conventional" --generate 3 --clipboard --exclude "*.json" --exclude "*.ts"

You can also integrate aicommit2 with Git via the prepare-commit-msg hook. This lets you use Git like you normally would, and edit the commit message before committing.

In the Git repository you want to install the hook in:

aicommit2 hook install

In the Git repository you want to uninstall the hook from:

aicommit2 hook uninstall

• READ: aicommit2 config get <key>
• SET: aicommit2 config set <key>=<value>

Example:

aicommit2 config get OPENAI
aicommit2 config get GEMINI.key
aicommit2 config set OPENAI.generate=3 GEMINI.temperature=0.5

1. Command-line arguments: use the format --[Model].[Key]=value

aicommit2 --OPENAI.locale="jp" --GEMINI.temperatue="0.5"

2. Configuration file: use INI format in the ~/.aicommit2 file or use set command. Example ~/.aicommit2:

# General Settings
logging=true
generate=2
temperature=1.0

# Model-Specific Settings
[OPENAI]
key="<your-api-key>"
temperature=0.8
generate=1
systemPromptPath="<your-prompt-path>"

[GEMINI]
key="<your-api-key>"
generate=5
includeBody=true

[OLLAMA]
temperature=0.7
model[]=llama3.2
model[]=codestral

The priority of settings is: Command-line Arguments > Model-Specific Settings > General Settings > Default Values.

The following settings can be applied to most models, but support may vary. Please check the documentation for each specific model to confirm which settings are supported.

👉 Tip: To set the General Settings for each model, use the following command.

aicommit2 config set OPENAI.locale="jp"
aicommit2 config set CODESTRAL.type="gitmoji"
aicommit2 config set GEMINI.includeBody=true

• Allow users to specify a custom system prompt

aicommit2 config set systemPrompt="Generate git commit message."

systemPrompt takes precedence over systemPromptPath and does not apply at the same time.

• Allow users to specify a custom file path for their own system prompt template
• Please see Custom Prompt Template

aicommit2 config set systemPromptPath="/path/to/user/prompt.txt"

• Files to exclude from AI analysis
• It is applied with the --exclude option of the CLI option. All files excluded through --exclude in CLI and exclude general setting.

aicommit2 config set exclude="*.ts"
aicommit2 config set exclude="*.ts,*.json"

NOTE: exclude option does not support per model. It is only supported by General Settings.

Default: conventional

Supported: conventional, gitmoji

The type of commit message to generate. Set this to "conventional" to generate commit messages that follow the Conventional Commits specification:

aicommit2 config set type="conventional"

Default: en

The locale to use for the generated commit messages. Consult the list of codes in: https://wikipedia.org/wiki/List_of_ISO_639_language_codes.

aicommit2 config set locale="jp"

Default: 1

The number of commit messages to generate to pick from.

Note, this will use more tokens as it generates more results.

aicommit2 config set generate=2

Default: true

Option that allows users to decide whether to generate a log file capturing the responses. The log files will be stored in the ~/.aicommit2_log directory(user's home).

• You can remove all logs below comamnd.

aicommit2 log removeAll

Default: false

This option determines whether the commit message includes body. If you want to include body in message, you can set it to true.

aicommit2 config set includeBody="true"

aicommit2 config set includeBody="false"

The maximum character length of the Subject of generated commit message

Default: 50

aicommit2 config set maxLength=100

The timeout for network requests in milliseconds.

Default: 10_000 (10 seconds)

aicommit2 config set timeout=20000 # 20s

The temperature (0.0-2.0) is used to control the randomness of the output

Default: 0.7

aicommit2 config set temperature=0.3

The maximum number of tokens that the AI models can generate.

Default: 1024

aicommit2 config set maxTokens=3000

Default: 0.9

Nucleus sampling, where the model considers the results of the tokens with top_p probability mass.

aicommit2 config set topP=0.2

Default: false

This option determines whether a specific model is enabled or disabled. If you want to disable a particular model, you can set this option to true.

To disable a model, use the following commands:

aicommit2 config set GEMINI.disabled="true"
aicommit2 config set GROQ.disabled="true"

Default: false

The codeReview parameter determines whether to include an automated code review in the process.

aicommit2 config set codeReview=true

NOTE: When enabled, aicommit2 will perform a code review before generating commit messages.

⚠️ CAUTION

• The codeReview feature is currently experimental.
• This feature performs a code review before generating commit messages.
• Using this feature will significantly increase the overall processing time.
• It may significantly impact performance and cost.
• The code review process consumes a large number of tokens.

• Allow users to specify a custom file path for code review

aicommit2 config set codeReviewPromptPath="/path/to/user/prompt.txt"

All AI support the following options in General Settings.

• systemPrompt, systemPromptPath, codeReview, codeReviewPromptPath, exclude, type, locale, generate, logging, includeBody, maxLength

Some models mentioned below are subject to change.

The OpenAI API key. You can retrieve it from OpenAI API Keys page.

aicommit2 config set OPENAI.key="your api key"

Default: gpt-4o-mini

The Chat Completions (/v1/chat/completions) model to use. Consult the list of models available in the OpenAI Documentation.

aicommit2 config set OPENAI.model=gpt-4o

Default: https://api.openai.com

The OpenAI URL. Both https and http protocols supported. It allows to run local OpenAI-compatible server.

aicommit2 config set OPENAI.url="<your-host>"

Default: /v1/chat/completions

The OpenAI Path.

Default: 0.9

The top_p parameter selects tokens whose combined probability meets a threshold. Please see detail.

aicommit2 config set OPENAI.topP=0.2

NOTE: If topP is less than 0, it does not deliver the top_p parameter to the request.

The Anthropic API key. To get started with Anthropic Claude, request access to their API at anthropic.com/earlyaccess.

Default: claude-3-haiku-20240307

Supported:

• claude-3-haiku-20240307
• claude-3-sonnet-20240229
• claude-3-opus-20240229
• claude-3-5-sonnet-20240620
• claude-3-5-sonnet-20241022

aicommit2 config set ANTHROPIC.model="claude-3-5-sonnet-20240620"

Anthropic does not support the following options in General Settings.

• timeout

The Gemini API key. If you don't have one, create a key in Google AI Studio.

aicommit2 config set GEMINI.key="your api key"

Default: gemini-1.5-pro

Supported:

• gemini-1.5-pro
• gemini-1.5-flash
• gemini-1.5-pro-exp-0801

aicommit2 config set GEMINI.model="gemini-1.5-pro-exp-0801"

Gemini does not support the following options in General Settings.

• timeout

The Mistral API key. If you don't have one, please sign up and subscribe in Mistral Console.

Default: mistral-tiny

Supported:

• open-mistral-7b
• mistral-tiny-2312
• mistral-tiny
• open-mixtral-8x7b
• mistral-small-2312
• mistral-small
• mistral-small-2402
• mistral-small-latest
• mistral-medium-latest
• mistral-medium-2312
• mistral-medium
• mistral-large-latest
• mistral-large-2402
• mistral-embed

The Codestral API key. If you don't have one, please sign up and subscribe in Mistral Console.

Default: codestral-latest

Supported:

• codestral-latest
• codestral-2405

aicommit2 config set CODESTRAL.model="codestral-2405"

The Cohere API key. If you don't have one, please sign up and get the API key in Cohere Dashboard.

Default: command

Supported models:

• command
• command-nightly
• command-light
• command-light-nightly

aicommit2 config set COHERE.model="command-nightly"

Cohere does not support the following options in General Settings.

• timeout

The Groq API key. If you don't have one, please sign up and get the API key in Groq Console.

Default: gemma2-9b-it

Supported:

• llama3-groq-70b-8192-tool-use-preview
• distil-whisper-large-v3-en
• llama3-70b-8192
• llama-3.2-11b-vision-preview
• whisper-large-v3-turbo
• gemma-7b-it
• llama3-groq-8b-8192-tool-use-preview
• llama-3.2-1b-preview
• llama-3.1-8b-instant
• mixtral-8x7b-32768
• llama-3.2-90b-text-preview
• llama3-8b-8192
• llama-guard-3-8b
• llama-3.2-90b-vision-preview
• llama-3.2-11b-text-preview
• llama-3.2-3b-preview
• llava-v1.5-7b-4096-preview
• whisper-large-v3
• llama-3.1-70b-versatile
• gemma2-9b-it

aicommit2 config set GROQ.model="llama3-8b-8192"

The Perplexity API key. If you don't have one, please sign up and get the API key in Perplexity

Default: llama-3.1-sonar-small-128k-chat

Supported:

• llama-3.1-sonar-small-128k-chat
• llama-3.1-sonar-large-128k-chat
• llama-3.1-sonar-large-128k-online
• llama-3.1-sonar-small-128k-online
• llama-3.1-8b-instruct
• llama-3.1-70b-instruct
• llama-3.1-8b
• llama-3.1-70b

The models mentioned above are subject to change.

aicommit2 config set PERPLEXITY.model="llama-3.1-70b"

The DeepSeek API key. If you don't have one, please sign up and subscribe in DeepSeek Platform.

Default: deepseek-coder

Supported:

• deepseek-coder
• deepseek-chat

aicommit2 config set DEEPSEEK.model="deepseek-chat"

The Huggingface Chat Cookie. Please check how to get cookie

# Please be cautious of Escape characters(\", \') in browser cookie string
aicommit2 config set HUGGINGFACE.cookie="your-cooke"

Default: CohereForAI/c4ai-command-r-plus

Supported:

• CohereForAI/c4ai-command-r-plus
• meta-llama/Meta-Llama-3-70B-Instruct
• HuggingFaceH4/zephyr-orpo-141b-A35b-v0.1
• mistralai/Mixtral-8x7B-Instruct-v0.1
• NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
• 01-ai/Yi-1.5-34B-Chat
• mistralai/Mistral-7B-Instruct-v0.2
• microsoft/Phi-3-mini-4k-instruct

aicommit2 config set HUGGINGFACE.model="mistralai/Mistral-7B-Instruct-v0.2"

Huggingface does not support the following options in General Settings.

• maxTokens
• timeout
• temperature
• topP

The Ollama Model. Please see a list of models available

aicommit2 config set OLLAMA.model="llama3.1"
aicommit2 config set OLLAMA.model="llama3,codellama" # for multiple models

aicommit2 config add OLLAMA.model="gemma2" # Only Ollama.model can be added.

OLLAMA.model is string array type to support multiple Ollama. Please see this section.

Default: http://localhost:11434

The Ollama host

aicommit2 config set OLLAMA.host=<host>

Not required. Use when your Ollama server requires authentication. Please see this issue.

aicommit2 config set OLLAMA.auth=<auth>

Not required. Use when your Ollama server requires authentication. Please see this issue.

aicommit2 config set OLLAMA.key=<key>

Few examples of authentication methods:

Default: 100_000 (100 seconds)

Request timeout for the Ollama.

aicommit2 config set OLLAMA.timeout=<timeout>

Ollama does not support the following options in General Settings.

• maxTokens

Check the installed version with:

aicommit2 --version

If it's not the latest version, run:

npm update -g aicommit2

aicommit2 supports custom prompt templates through the systemPromptPath option. This feature allows you to define your own prompt structure, giving you more control over the commit message generation process.

To use a custom prompt template, specify the path to your template file when running the tool:

aicommit2 config set systemPromptPath="/path/to/user/prompt.txt"
aicommit2 config set OPENAI.systemPromptPath="/path/to/another-prompt.txt"

For the above command, OpenAI uses the prompt in the another-prompt.txt file, and the rest of the model uses prompt.txt.

NOTE: For the systemPromptPath option, set the template path, not the template content

Your custom template can include placeholders for various commit options. Use curly braces {} to denote these placeholders for options. The following placeholders are supported:

• {locale}: The language for the commit message (string)
• {maxLength}: The maximum length for the commit message (number)
• {type}: The type of the commit message (conventional or gitmoji)
• {generate}: The number of commit messages to generate (number)

Here's an example of how your custom template might look:

Generate a {type} commit message in {locale}.
The message should not exceed {maxLength} characters.
Please provide {generate} messages.

Remember to follow these guidelines:
1. Use the imperative mood
2. Be concise and clear
3. Explain the 'why' behind the change

Please note that the following text will ALWAYS be appended to the end of your custom prompt:

Lastly, Provide your response as a JSON array containing exactly {generate} object, each with the following keys:
- "subject": The main commit message using the {type} style. It should be a concise summary of the changes.
- "body": An optional detailed explanation of the changes. If not needed, use an empty string.
- "footer": An optional footer for metadata like BREAKING CHANGES. If not needed, use an empty string.
The array must always contain {generate} element, no more and no less.
Example response format:
[
  {
    "subject": "fix: fix bug in user authentication process",
    "body": "- Update login function to handle edge cases\n- Add additional error logging for debugging",
    "footer": ""
  }
]
Ensure you generate exactly {generate} commit message, even if it requires creating slightly varied versions for similar changes.
The response should be valid JSON that can be parsed without errors.

This ensures that the output is consistently formatted as a JSON array, regardless of the custom template used.

You can load and make simultaneous requests to multiple models using Ollama's experimental feature, the OLLAMA_MAX_LOADED_MODELS option.

• OLLAMA_MAX_LOADED_MODELS: Load multiple models simultaneously

Follow these steps to set up and utilize multiple models simultaneously:

First, launch the Ollama server with the OLLAMA_MAX_LOADED_MODELS environment variable set. This variable specifies the maximum number of models to be loaded simultaneously. For example, to load up to 3 models, use the following command:

OLLAMA_MAX_LOADED_MODELS=3 ollama serve

Refer to configuration for detailed instructions.

Next, set up aicommit2 to specify multiple models. You can assign a list of models, separated by commas(,), to the OLLAMA.model environment variable. Here's how you do it:

aicommit2 config set OLLAMA.model="mistral,dolphin-llama3"

With this command, aicommit2 is instructed to utilize both the "mistral" and "dolphin-llama3" models when making requests to the Ollama server.

aicommit2

Note that this feature is available starting from Ollama version 0.1.33 and aicommit2 version 1.9.5.

• Login to the site you want
• You can get cookie from the browser's developer tools network tab
• See for any requests check out the Cookie, Copy whole value
• Check below image for the format of cookie

When setting cookies with long string values, ensure to escape characters like ", ', and others properly.

• For double quotes ("), use \"
• For single quotes ('), use \'

This project uses functionalities from external APIs but is not officially affiliated with or endorsed by their providers. Users are responsible for complying with API terms, rate limits, and policies.

For bug fixes or feature implementations, please check the Contribution Guide.

Thanks goes to these wonderful people (emoji key):

@eltociear 📖

@ubranch 💻

@bhodrolok 💻

@ryicoh 💻

@noamsto 💻

If this project has been helpful, please consider giving it a Star ⭐️!

Maintainer: @tak-bro