Turn a plain English request like 'list all pdf files' into the correct shell command for your OS.
Get an automatic fix suggestion when a generated command fails to run.
Ask for a line-by-line explanation of an unfamiliar terminal command before running it.
Work fully offline since command generation happens through a local AI model, not a cloud service.
| moisesfilho/shelloma | 12vault/ravel | alexremn/finalizer-doctor | |
|---|---|---|---|
| Stars | 3 | 3 | 3 |
| Language | Go | Go | Go |
| Setup difficulty | moderate | easy | easy |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | developer | developer | ops devops |
Figures from each repo's GitHub metadata at analysis time.
Requires Ollama installed and running locally with a coding-focused model downloaded.
Shelloma is a command line tool that turns plain English instructions into the actual terminal commands you would otherwise have to remember or look up. You type something like list all pdf files in downloads folder, and Shelloma sends that instruction, along with details about your operating system and current folder, to a locally running AI model through a program called Ollama, which then writes the exact Bash, Zsh, PowerShell, or CMD command that would do what you asked. Before anything runs, Shelloma shows you the generated command in a small card and lets you choose what to do with it: run it right away, ask for a line by line explanation of what it does, edit it first, copy it to your clipboard, or cancel. If a command fails once you run it, Shelloma looks at the error output, works out what likely went wrong, and suggests a fix you can then choose to run. Because everything happens through Ollama, which runs AI models directly on your own computer, no instructions or data are sent to an outside server, and it works even without an internet connection once the model is downloaded. Shelloma also speaks English, Brazilian Portuguese, and Spanish out of the box, and it will notice if Ollama itself is not running and offer to start it for you. It is written in Go and ships as a single downloadable file for Linux, macOS, and Windows, with packages available for common Linux package managers as well as a portable AppImage. Setup requires installing Ollama separately and downloading a small AI model suited for coding tasks, such as Qwen 2.5 Coder, before Shelloma can start generating commands. The project is released under the MIT license.
A command line tool that converts plain English instructions into ready-to-run terminal commands using a locally running AI model via Ollama.
Mainly Go. The stack also includes Go, Ollama.
MIT license: use, modify, and share freely, including commercially, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.