An AI-based CLI assistant to help you with command line & shell.
Project description
how-cli
An AI-based CLI assistant to help you with command line & shell.
Demo
https://github.com/user-attachments/assets/effefe1a-c0ed-4b60-838c-98f992f6c25f
Installation
1. Using pip
- Ensure python is installed on your system. (Tested against Python 3.11+)
- Install the package using pip.
pip install -U how-cli
2. Manual Installation
- Clone the repository & cd into it
git clone https://github.com/FireHead90544/how-cli && cd how-cli
- Ensure you're in a virtual environment.
- Install the application.
pip install -e .
Usage
$ how [OPTIONS] COMMAND [ARGS]...
Options:
-v, --version
: Shows the version of the application--install-completion
: Install completion for the current shell.--show-completion
: Show completion for the current shell, to copy it or customize the installation.--help
: Show this message and exit.
Commands:
setup
: Sets up the configuration required to run...to
: Sends the task to the LLM for analysis.
how setup
Sets up the configuration required to run the application. Set the LLM Provider & the corresponding API Key.
Usage:
$ how setup [OPTIONS]
Options:
--interactive / --no-interactive
: Whether to use interactive mode for setting up configuration? [default: interactive]--provider TEXT
: The LLM Provider, needs to be passed explicitly if using --no-interactive mode.--api-key TEXT
: The API Key for the LLM provider, needs to be passed explicitly if using --no-interactive mode.--help
: Show this message and exit.
how to
Sends the task to the LLM for analysis. Returns the commands to be executed in order to achieve that.
Usage:
$ how to [OPTIONS] TASK
Arguments:
TASK
: The command line task to perform. [required]
Options:
--help
: Show this message and exit.
Providers
how-cli
uses ChatModels as they support chat messages as opposed to TextModels and below model providers and their corresponding models are available to use. If you could test the models that are marked as ❌, please create an issue or pull request along with the test results.
Provider | Model | Package | Tested |
---|---|---|---|
GoogleGenAI | gemini-1.5-flash |
langchain-google-genai |
✅ |
GoogleVertexAI | gemini-1.5-flash |
langchain-google-vertexai |
❌ |
GroqMistralAI | mixtral-8x7b-32768 |
langchain-groq |
✅ |
GroqLLaMa | llama3-70b-8192 |
langchain-groq |
✅ |
OpenAI | gpt-4o |
langchain-openai |
❌ |
Anthropic | claude-3-5-sonnet-20240620 |
langchain-anthropic |
❌ |
License
how-cil
is licensed under the MIT License, it can be found here.
Honourable Mentions
This project is greatly inspired by kynnyhsap's how. Though my implementation is completely different (refer to the below image for architectural details), but at the core both the projects aims to do the same thing. Also, check out LangChain & Typer using which this project was built.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file how_cli-0.2.0.tar.gz
.
File metadata
- Download URL: how_cli-0.2.0.tar.gz
- Upload date:
- Size: 9.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.11.9
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 22a8676e273360ac65f00ef7a00399698e661957956269da2452f6a7e2bf67ad |
|
MD5 | cb32181d289a6116c16a3633f88b8653 |
|
BLAKE2b-256 | 78c6d5a4b949ba6bd75594e88926462f66b7e33111edb0c87e67c13fb1005026 |
File details
Details for the file how_cli-0.2.0-py3-none-any.whl
.
File metadata
- Download URL: how_cli-0.2.0-py3-none-any.whl
- Upload date:
- Size: 10.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.11.9
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 894d071a76d21f01d67ac7ab51d6e7f031986a6feee0fc3d12397ae740a513ea |
|
MD5 | 176c688ad8b45e5be3ed5a533678f7b3 |
|
BLAKE2b-256 | 90ae302774407f3489500fecdf94fa9f1a393110dbcd1dec0a08f675b55431d1 |