Install

Download Elissa from the latest release. One version is published for every system, and installed copies update themselves.

Windows macOS Linux

Requirements: Windows 10 (1809 or later) or Windows 11, 64-bit, with WebView2 (built into Windows 11), a microphone and speakers.

1. Download Elissa_x.y.z_x64-setup.exe.
2. Run it. No administrator rights are needed; Elissa installs for your user only.
3. Elissa opens in the bottom-right corner and asks you to set up your AI.

SmartScreen or Defender warnings

The installer isn't code-signed yet, so Windows SmartScreen may warn you the first time. Choose More info → Run anyway.

If Microsoft Defender flags it, that's a false positive common with unsigned apps built with PyInstaller. Update Defender (Update-MpSignature in PowerShell) and try again, or report it. Every release is built and self-tested by GitHub Actions from the public code, and its updates are signed. See the code signing policy.

Requirements: macOS 11 or later on Apple Silicon (M1 or newer). There is no Intel build yet.

1. Download Elissa_x.y.z_aarch64.dmg, open it and drag Elissa to Applications.
2. The app isn't notarized by Apple, so the first time macOS won't open it. Go to System Settings → Privacy & Security, scroll down and choose Open Anyway. Or, in Terminal:

   xattr -dr com.apple.quarantine /Applications/Elissa.app
   

3. Allow the microphone when she asks. After an update macOS may ask again; Elissa tells you where to turn it back on.
4. The first time she controls Spotify or Music, macOS asks you to let Elissa control that app. Allow it.
5. To let her use other windows, allow Elissa in System Settings → Privacy & Security → Accessibility when she asks.

Ubuntu, Debian and derivatives: download Elissa_x.y.z_amd64.deb and install it:

sudo apt install ./Elissa_x.y.z_amd64.deb

Any other distribution: download the .AppImage, make it executable and run it. It needs PortAudio for the microphone (libportaudio2 on Debian-based systems):

chmod +x Elissa_*.AppImage
./Elissa_*.AppImage

For music turned down while you talk, the system volume and offline voices, also install:

sudo apt install pulseaudio-utils espeak-ng

To add folders from Settings → Folders she can use, install zenity. On Wayland her eyes don't follow the cursor outside her window.

After installing

Elissa appears as a small character in the bottom-right corner, and her icon sits in the tray. Drag her anywhere. Continue with the first steps.

What works on each system

macOS and Linux are previews. What isn't available on your system doesn't show up in Settings.

Feature	Windows	macOS	Linux
Talk, chat, files, web, memory, reminders, email
Speech recognition in the cloud or on your computer
"Hey Elissa" wake phrase
Natural voices (Edge) and offline voices			¹
Spotify and MCP integrations
Open installed apps
"What's playing?", pause, skip	any player	Spotify and Music	any MPRIS player
Turn music down while you talk, system volume			¹
Terminal commands in your projects	PowerShell	zsh	bash
Claude Code and Codex alerts
Use other windows (read, click, type)		²	—

¹ With pulseaudio-utils (also for PipeWire) and espeak-ng installed. ² macOS asks for the Accessibility permission the first time.

Updates

Elissa checks for a new version 20 seconds after starting and every 6 hours. When there is one, the tray menu shows Instalar la versión X and Settings shows Update now. Updates are signed and verified before they are installed. You can also check by hand from the tray: Buscar actualizaciones.

Uninstall

• Windows: Settings → Apps → Installed apps → Elissa → Uninstall.
• macOS: quit Elissa from the tray and move it from Applications to the Trash.
• Linux: sudo apt remove elissa, or delete the AppImage.

Uninstalling doesn't delete your settings, memory and logs. To remove them too, delete the data folder, and remove the Elissa entries from your system's keychain.