Whasp

Visual AI workflow studio

Download for Windows v0.1.0 · 12.8 MB Website ↗

Windows

Get it

Whasp-0.1.0.msi

12.8 MB · SHA-256 d24f7c0e7875…

Download

If you've already added the repository, just run:

sudo apt install whasp

First-time setup (add repository)

curl -fsSL -o /tmp/colaapps-archive-keyring.gpg \
     https://colaapps.com/repo/colaapps-archive-keyring.gpg &&
sudo install -D -m 0644 /tmp/colaapps-archive-keyring.gpg \
     /etc/apt/keyrings/colaapps-archive-keyring.gpg

sudo tee /etc/apt/sources.list.d/colaapps.list > /dev/null <<'EOF'
deb [signed-by=/etc/apt/keyrings/colaapps-archive-keyring.gpg] https://colaapps.com/repo stable main
EOF

sudo apt update
sudo apt install whasp

The first two commands install the ColaApps signing key, so apt verifies every index it downloads. Needs curl. Debian and Ubuntu.

Added ColaApps before, with [trusted=yes]? That turns signature checking off — run the block above again, it overwrites the old line.

About

Whasp — visual AI workflow studio

Design AI work on a node canvas — connect HTTP requests, LLM prompts, and logic into pipelines, then run them. Two ways to use it:

Local — runs fully on your machine. Point it at your own EWACS / Ollama endpoint in Settings; nothing leaves your computer and there are no credits. Workflows save locally and export/import as portable .whasp.json files.
Remote — connect to a Whasp server (e.g. your whasp.thework.cloud tenancy), sign in with DoIt, and use its models, saved workflows, and shared library.

Build a workflow offline, then import the same .whasp.json into your cloud tenancy to publish it.

Under the hood

A native shell on our own stack — a Rust window hosting the system WebView (no bundled browser), with the Whasp backend running in-process. Small and fast: one process, ~13 MB installer.

Install

Windows 64-bit. The installer is not yet code-signed, so Windows SmartScreen may warn on first run — choose More info -> Run anyway.