A shell interface for MLX models
Project description
querysh
A command-line interface that uses local, offline AI to help you write and execute shell commands. Powered by MLX-optimized Llama 3.2 1B Instruct model running entirely on your machine.
Features
- Natural language to shell command conversion
- Command validation and safety checks
- Interactive command execution
- Rich terminal interface
- Support for command suggestions and modifications
- Fully local and offline operation using MLX-optimized Llama 3.2 1B Instruct model
- No internet connection or API keys required
Installation
Requirements
- Python 3.8 or higher
- macOS (for MLX support)
Dependencies
- mlx-lm>=0.0.6
- rich>=13.7.0
Install from PyPI
pip install querysh
Install from source
git clone https://github.com/pabl0cesar/querysh.git
cd querysh
pip install -e .
Usage
Simply run the querysh command in your terminal:
querysh
Then describe what you want to do in natural language. For example:
- "List all files in the current directory"
- "Find all Python files modified in the last week"
- "Create a backup of my documents folder"
The AI will suggest appropriate shell commands, and you can choose to execute them or modify them.
Development
Setup
- Clone the repository:
git clone https://github.com/pabl0cesar/querysh.git
cd querysh
- Create and activate a virtual environment:
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
- Install development dependencies:
pip install -e ".[dev]"
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
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file querysh-0.1.0.tar.gz.
File metadata
- Download URL: querysh-0.1.0.tar.gz
- Upload date:
- Size: 8.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
140defcb7652e9bbcf00849ae8dbfb854f43dc200fd2764764f6f355294e6303
|
|
| MD5 |
823e7298bcd327f750ba7724c0073dd6
|
|
| BLAKE2b-256 |
6dc5359ee9551e56d2983eb8c9f61af3b5ad3b11be15b6763041ec93251966e5
|
File details
Details for the file querysh-0.1.0-py3-none-any.whl.
File metadata
- Download URL: querysh-0.1.0-py3-none-any.whl
- Upload date:
- Size: 9.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
d14920b14f8a38d800a66ff25e51506c4a9994ae13325d36fc082121914a65f8
|
|
| MD5 |
e46032543dbabbd8bcfc1d81fd39648c
|
|
| BLAKE2b-256 |
6e5d1fe5fd5fe5f6922cdc2762b9cfc7b2fa7aec79efad2454881ff25f42378d
|