agentstead.dev Docs
Everything you need to deploy, configure, and manage AI agent runtimes on agentstead.dev.
Welcome to the agentstead.dev docs. Pick a runtime, hit deploy, and you get a full Linux desktop for your AI agent — persistent, always-on, and ready to use from your browser.
First time? Start with Getting Started — you'll have an agent online in around 90 seconds.
Getting started
Getting Started
Sign in, deploy, and open your first agent desktop.
Choose Your Runtime
OpenClaw, OpenFang, NullClaw, or ZeroClaw — which is right for you?
Pre-installed Tools
Claude Code, Codex, OpenCode, Neovim, Homebrew and more — all pre-installed.
Manage your agents
Agent Lifecycle
Deploy, restart, redeploy, stop, and destroy agents.
Networking & SSH
Public URLs, Tailscale private access, and SSH.
Data & Backups
Persistent workspaces, on-demand backups, and scheduled backups.
Schedules
Automate recurring backups and restarts with cron schedules.
Runtime deep-dives
OpenClaw
Gateway + full desktop. Best for Claude Code, Codex, and general AI coding.
OpenFang
Rust API runtime with Ollama fallback and ffmpeg.
NullClaw
Zig-based runtime with a minimal footprint.
ZeroClaw
Rust runtime with pairing-code desktop auth.
API
API Overview
Base URL, authentication, scopes, and error format.
Personal Access Tokens
Create and manage PATs for scripted API access.
API Reference
Endpoint reference for agents, secrets, backups, schedules, and tokens.