Chat with LLM in your terminal, be it shell generator, story teller, linux-terminal, etc.

pip install -U shgpt

This will install two commands: sg and shgpt, which are identical.

After install, use sg --init to create required directories(mainly ~/.shellgpt).

ShellGPT has three modes to use:

• Direct mode, sg [question] or pipeline like echo question | sg.
• TUI mode, sg, tailored for infer shell command.
• REPL mode, sg -l, chat with LLM.

See conf.py for detailed configs.

There are some key bindings to use in TUI: option

• ctrl+j, Infer answer
• ctrl+r, Run command
• ctrl+y, Yank command

There are 4 built-in system role contents in shellgpt:

• default, used for ask general questions
• code, used for ask programming questions
• shell, used for infer shell command
• cm, used for generate git commit message, like git diff | sg -r cm

Users can define their own content in ~/.shellgpt/roles.json, it a JSON map with

• key being role name and
• value being role content

Or you can just copy roles.json to play with, it's generated from Awesome ChatGPT Prompts.

$ shgpt -r linux-terminal pwd
/home/user

$ shgpt -r javascript-console 0.1 + 0.2
0.3

• Ollama, you need to download models before try shellgpt.

GPL-3.0