kash v2.2.1

What's Changed

Fixed

• Chat answers ignored the corpus behind a Gemini gateway.
  /v1/chat/completions sent the agent prompt and the retrieved context as two
  separate system messages, and Gemini behind an OpenAI-compatible gateway
  keeps only the first. The context was dropped without an error, so answers
  came from the model's own priors even when retrieval had found the right
  passage: asked what the Trishul represents, an agent built from a book that
  answers it in one sentence returned generic lore and no citation. The prompt
  and the context now travel in a single system message. No rebuild needed.

Installation

Linux (amd64)

curl -L https://github.com/akashicode/kash/releases/download/v2.2.1/kash_linux_amd64.tar.gz | tar -xz
sudo mv kash /usr/local/bin/

Linux (arm64)

curl -L https://github.com/akashicode/kash/releases/download/v2.2.1/kash_linux_arm64.tar.gz | tar -xz
sudo mv kash /usr/local/bin/

macOS (Apple Silicon)

curl -L https://github.com/akashicode/kash/releases/download/v2.2.1/kash_darwin_arm64.tar.gz | tar -xz
sudo mv kash /usr/local/bin/

macOS (Intel)

curl -L https://github.com/akashicode/kash/releases/download/v2.2.1/kash_darwin_amd64.tar.gz | tar -xz
sudo mv kash /usr/local/bin/

Windows (amd64)

Download kash_windows_amd64.zip, extract, and add kash.exe to a directory on your PATH.

---

For full usage instructions, configuration options, and examples, see the README.