Quick Start
-
1
Install the KDCube CLI
# Recommended: pipx (isolated installation) pipx install kdcube-cli # Or with pip pip install kdcube-cliPrerequisites: Python 3.11+, Docker, Git.
-
2
Run the setup wizard
kdcubeThe wizard is fully interactive — you can enter everything manually, or point it at deployment descriptors and it applies them automatically:
assembly.yaml·gateway.yaml·bundles.yaml. For prepared descriptors,kdcube --descriptors-location /path/to/descriptors --latestuses the latest released platform,kdcube --descriptors-location /path/to/descriptors --release 2026.4.11.012pins a specific release, andkdcube --descriptors-location /path/to/descriptors --build --upstreambuilds from the latest upstream repo state. Creates~/.kdcube/kdcube-runtime/, pulls images or builds locally, and starts Docker Compose. Secrets are injected into the kdcube-secrets service at setup time — never written to disk. -
3
Open the UI
http://localhost:${KDCUBE_UI_PORT}/chatbot/chatYou're now running a full KDCube stack with the built-in example bundle set.
Setup Flow
Workdir Layout
~/.kdcube/kdcube-runtime/
├─ config/
│ ├─ .env # Base Docker Compose env
│ ├─ assembly.yaml # Platform version, auth type, domain
│ ├─ bundles.yaml # Bundle definitions ← edit this
│ ├─ gateway.yaml # Rate limits, circuit breaker
│ ├─ install-meta.json # CLI install metadata
│ ├─ frontend.config.*.json # UI auth config
│ ├─ nginx_proxy.conf # Nginx reverse proxy
│ └─ nginx_ui.conf # Nginx UI config
├─ data/
│ ├─ bundle-storage/ # Per-bundle persistent storage
│ ├─ bundles/ # Local path bundle root
│ ├─ git-bundles/ # Optional git bundle cache root
│ ├─ exec-workspace/ # Code execution sandboxes
│ ├─ kdcube-storage/ # Conversations, artifacts, files
│ ├─ nginx/ # Nginx runtime data
│ ├─ postgres/ # Database volume
│ └─ redis/ # Redis persistence
└─ logs/ # Service logs
Secrets are never stored on disk — they are injected into kdcube-secrets at setup time and resolved in-memory at runtime.
Full CLI reference: kdcube_cli/README.md