LaunchKit · 2026
Back to Skills

agent-spawner

Spawn a new OpenClaw agent through conversation.

0
575 downloads
by @austineral

Setup & Installation

openclaw skills install @austineral/agent-spawner

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install agent-spawner

What This Skill Does

Deploys a new OpenClaw agent through a short conversation. Reads the current agent's config silently and carries over the API key, model, plugins, and skills automatically. The user answers 2-3 questions and never edits a file.

Automates config carryover so you don't have to manually track and copy API keys, plugin specs, and skill files across agents.

When to use it

  • Spinning up a second agent for an isolated project
  • Cloning your current agent setup to a remote server
  • Provisioning a fresh agent for a teammate with your existing config
  • Creating a dedicated agent for a specific workflow
  • Deploying a test agent without touching production config

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Spin up a new agent called 'research-bot' on Docker locally

AGENT
  1. 1Read current agent config, env vars, installed plugins, and workspace skills
  2. 2Ask where to deploy, confirm the agent name, and prompt for any special purpose
  3. 3Present full deploy plan showing every carried-over item before touching anything
  4. 4Run Docker onboard with detected provider flags, then bring up the gateway container
  5. 5Copy plugins and skills to the new agent, restart the gateway, and read the generated token
OUTPUT

Running agent at http://localhost:18789/ with gateway token, ready to receive its first message

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

Docker (for Docker-based deployment)