Getting Started

Tabbify gives your AI an office. You describe an outcome in chat, your AI writes the code, and Tabbify runs it as a real service — with containers, domains, private networking, secrets, and an audit trail. This page is the orientation; the Quickstart is the hands-on deploy.

The model

• Office — the namespace you own: its apps, services, secrets, and approvals. Chat is the control plane; everything below is runtime.
• App — one workload identified by a UUID v7. You publish it with the CLI, and a single tabbify.toml is the source of truth for build, runtime, and deploy targets.
• Approvals — capabilities like telegram:send or domain:connect are requested before anything acts, and recorded in an append-only event log. Secrets never reach the prompt — they live in an encrypted store and are injected as environment. (The chat approval card is designed but not yet wired in the frontend.)

The shortest path

Three steps move an app from your disk to a URL: package it, start it on a supervisor, reach it through the node.

# 1. Package and upload (mints a UUID on first push, recorded in tabbify.lock)
tcli push examples/hello-tabbify

# 2. Start it on a supervisor in the mesh
curl -X POST -H 'Authorization: Bearer tabbify-dev-node-key' \
  http://3.124.69.92:8090/v1/apps/<uuid>/start

# 3. Reach it through the node
curl http://3.124.69.92:8090/app/<uuid> \
  -H 'Authorization: Bearer tabbify-dev-node-key'
# 200 OK  "Hello, Tabbify!"

The node runs on port 8090 (auth occupies 8080 on the same host). tabbify-dev-node-key is the hardcoded RnD default — set TABBIFY_NODE_KEY to a long random secret for any shared deployment. Routing is IPv6 ULA over WireGuard, so a supervisor needs NET_ADMIN and /dev/net/tun.

The full walkthrough — what each command prints, where artifacts land, how to stop and warm-restart — is in the Quickstart.

Next steps

• Core concepts — the mental model behind offices, apps, and the mesh.
• The tabbify.toml manifest — the unified config format.
• Runtimes — WASM, Firecracker, and Docker, chosen per deploy target.
• The deploy pipeline — deploy straight from a git push.
• The Node API — the REST + MCP surface for driving apps.
• CLI reference — everything tcli does.
• Self-hosting a node — run your own supervisor.