install

Download & Install

ChapterDesk runs on Linux and Windows as a desktop app. It starts its own local server automatically — you do not need to set up anything else.

Linux (AppImage / .deb / .rpm)

1. Go to the Releases page.
2. Download one of:
   • .AppImage — run anywhere, no installation
   • .deb — Debian / Ubuntu / Mint …
   • .rpm — Fedora / RHEL / openSUSE …

Make the AppImage executable and start it:

chmod +x ChapterDesk-*.AppImage
./ChapterDesk-*.AppImage

Or install with your distro tooling:

# Debian/Ubuntu
sudo apt install ./ChapterDesk-*.deb

# Fedora
sudo dnf install ./ChapterDesk-*.rpm

Your data is stored in:

~/.config/ChapterDesk/data/chapterdesk.sqlite

Windows

1. Go to the Releases page.
2. Download the ChapterDesk Setup *.exe installer (NSIS).
3. Run it — you can choose the install folder.
4. Start ChapterDesk from the Start menu or the desktop shortcut.

Your data is stored in:

%APPDATA%\ChapterDesk\data\chapterdesk.sqlite

If Windows Defender SmartScreen warns about an unknown publisher, choose More info → Run anyway. The build is unsigned unless a code-signing certificate is added.

What you need for AI

ChapterDesk does not ship a model. You bring a provider that speaks the OpenAI-style chat API — OpenRouter is the usual choice (one key, many models). A long-context, instruction-following model is enough; you are giving scene directions, not running a toolbox agent.

See Settings for connecting a provider and choosing a model.

Build it yourself (developers)

git clone https://github.com/aschoerk/chapterdesk.git
cd chapterdesk
npm install
npm install --prefix chapterdesk-api-server   # bundled API server deps
npm run electron:build                        # AppImage / .deb / .rpm / .exe

For daily development:

npm run electron:dev   # Angular dev server + Electron + local chat-server