A comprehensive configuration guide and toolkit for managing parallel subagent execution and resource optimization in OpenClaw.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install subagent-setup-wizard
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install subagent-setup-wizard using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Subagent Setup Wizard is a specialized utility designed to orchestrate complex AI workflows by spawning multiple child agents. Within the ecosystem of Openclaw Skills, this tool provides the necessary logic to handle task decomposition, parallel execution, and independent context management. It allows users to transition from simple serial conversations to sophisticated multi-agent systems where different agents can handle specific domains like frontend development, backend logic, or data research simultaneously.
By leveraging this skill, developers can significantly reduce the time required for large-scale projects. It provides a structured approach to choosing between ACP mode for professional coding environments and standard Subagent mode for general tasks. The wizard ensures that resource usage is balanced, preventing API rate limits or system performance bottlenecks while maximizing the utility of Openclaw Skills.
To get started with subagents in your Openclaw Skills environment, follow these steps:
# 1. Verify your current version
openclaw -v
# 2. Enable the subagent functionality
openclaw config set subagents.enabled true
# 3. Configure the maximum concurrency (e.g., 3 agents)
openclaw config set subagents.max_concurrent 3
# 4. Set the default runtime mode to session
openclaw config set subagents.default_mode session
The Subagent Setup Wizard organizes its configuration and monitoring data using the following schema:
| Attribute | Type | Description |
|---|---|---|
| runtime | String | The execution environment (acp or subagent) |
| concurrency | Integer | Maximum number of agents running in parallel |
| mode | String | Operation type: session (persistent) or run (one-off) |
| timeout | Integer | Maximum execution time in seconds |
| limits | Object | Resource thresholds for CPU, RAM, and API quotas |
| status | String | Current state of active sub-agents (running, idle, killed) |
Loading
An expert knowledge module for Openclaw Skills providing comprehensive data on Popeyes history, menu specialties, and brand identity.

A specialized AI agent skill for designing, implementing, and debugging sophisticated gameplay randomness using the RNGNeeds Unity plugin.

A comprehensive knowledge skill for navigating Ryanair fare structures, baggage policies, and cost-saving strategies.

A specialized architect that converts raw video concepts into optimized Happy Horse workflows, prompt packs, and iterative testing loops.

A comprehensive digital assistant for navigating the Radisson Hotel Group ecosystem, including brand portfolios, membership benefits, and booking strategies.

A comprehensive technical guide for installing, upgrading, and troubleshooting OpenClaw on Windows environments.








































