Skip to content

Welcome to AgentDock

AgentDock is a desktop app that gives you one unified chat interface for the coding-agent CLIs you already have installed.

AgentDock is a native desktop app (Electron + React) that puts Claude Code, OpenAI Codex, and Google Antigravity behind one consistent chat interface. Point it at a project folder, pick an agent, and you get rich Markdown rendering, syntax-highlighted diffs, and structured tool/approval activity instead of a raw terminal.

AgentDock doesn't provide model access, accounts, or subscriptions of its own — it's a shell around coding-agent CLIs you install and authenticate yourself. Everything runs locally: sessions are stored in a local SQLite database, and your code, prompts, and credentials are never transmitted anywhere by AgentDock.

What's in these docs

  • Getting Started — install AgentDock, connect an agent CLI, and start your first session.
  • Agents — how AgentDock detects and talks to Claude Code, Codex, and Antigravity.
  • Workspaces — projects, sessions, and how state persists between launches.
  • AIgency

    AgentDock Pro

    — multi-agent task orchestration, available in AgentDock Pro.
  • Compare Mode — run the same prompt across multiple agents side by side.
  • Troubleshooting — fixes for the most common setup problems.

Free vs. Pro

Most of this documentation applies to both AgentDock and AgentDock Pro. Pages and sections that only apply to Pro are marked with a Pro badge. See AgentDock vs. AgentDock Pro for the full picture.

Where to start

If this is your first time installing AgentDock, start with Installation, then Connecting your coding agents and Creating your first workspace.

If you already have AgentDock running, Core concepts is a quick way to get oriented on the vocabulary the rest of these docs use (workspaces, sessions, agents, and — for Pro users — runs and tasks).