Install

macOS and Linux, Intel and ARM. Downloads the binary for your platform into ~/.local/bin, verifies it against the release checksums, and tells you if that directory is not on your PATH.

curl -fsSL https://raw.githubusercontent.com/devjasha/raunen/main/install.sh | sh

Nothing is needed at runtime — one static binary, about 3 MB compressed.

Changing where it goes

Put the variables on the sh side of the pipe. In VAR=x curl … | sh they would apply to curl, not to the shell running the script:

shell

$ curl -fsSL https://raw.githubusercontent.com/devjasha/\
    raunen/main/install.sh \
  | RAUNEN_INSTALL_DIR=/usr/local/bin RAUNEN_VERSION=v0.1.0 sh

Piping a script into a shell is worth reading first: curl -fsSL …/install.sh | less.

From source

Needs Go 1.25 or newer.

shell

$ go build -o ~/.local/bin/raunen .

Then have a model to talk to

Anything speaking the OpenAI API will do. With Ollama:

shell

$ ollama pull qwen3.5
$ cd ~/Projects/my-thing
$ raunen

raunen: no default model set, using ollama/qwen3.5:latest

The first run writes ~/.config/raunen/config.json and, since no model is configured yet, asks your endpoints what they have and picks one. Nothing here assumes a particular model is installed.