Skip to main content

​
Local-first architecture

Avibe runs on your machine. Your codebase, working directory, agent CLI, agent process, logs, and local configuration stay local. Local-first is an architectural stance, not a setting — see Design Philosophy.

​
Data flow

You
  -> chat app
  -> Avibe on your machine
  -> Claude Code / OpenCode / Codex
  -> Avibe
  -> chat app
  -> You
Avibe is the local runtime that lets your AI colleague live in chat while the actual agent work stays on your machine. It does not add a hidden LLM loop between you and your chosen agent.

​
Why this matters

  • No hosted coding VM is required.
  • No public inbound port is required for normal chat control.
  • The agent sees the same local project state you would see in a terminal.
  • Token overhead stays near zero because Avibe does not act as a second reasoning agent in front of your chosen agent.

​
What leaves your machine

To be precise about the boundary: the only thing that leaves is what your chosen agent sends to its AI provider — the prompt and context for the task, going to Anthropic, OpenAI, or whichever provider you configured. Your code at rest, your keys, your logs, and the agent process never leave. When you use avibe.bot, it carries only the control-plane handshake and the tunnel, never your data.

​
What avibe.bot is for

avibe.bot is the hosted control plane for secure remote Web UI access. When you run: 
vibe remote
Avibe helps pair your local Web UI with an authenticated remote URL. Browser traffic goes through the tunnel to your local machine; your data and the agent runtime remain local. Access is fail-closed — auth, routing, and host checks default to deny — so a misconfiguration locks you out rather than exposing your machine.