openclaw-multi-agent

Configure OpenClaw multi-agent deployments with four specialized agents.

1|2|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/hanlinlibham/skills --skill openclaw-multi-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openclaw-multi-agent
Source: https://github.com/hanlinlibham/skills/tree/main/openclaw-adj-skill
Command: npx skills add https://github.com/hanlinlibham/skills --skill openclaw-multi-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Configure and manage OpenClaw multi-agent deployments with four specialized agents (Shandian, Work, Research, Coding), including agent-to-agent communication, tool permissions, sandbox configuration, workspace setup, and permission management.

Core Features & Use Cases

  • Architect and deploy a four-agent OpenClaw setup with defined roles, models, and tool access
  • Generate and apply workspace templates, memory files, and agent configurations
  • Troubleshoot connectivity, permissions, and sandbox issues across agents

Quick Start

Run the provided setup script to initialize the agent directories and memory templates.

Frequently Asked Questions about openclaw-multi-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure multi-agent deployment roles and tool permissions in OpenClaw?

To configure multi-agent deployment in OpenClaw, you define specialized agent roles, set tool access permissions, and apply workspace templates for the four-agent setup. The provided setup script initializes agent directories and memory templates to automate this configuration.

What is the best way to set up inter-agent communication and sandbox settings for a multi-agent workspace?

Setting up inter-agent communication and sandbox settings requires validating agent-to-agent networking and applying predefined sandbox configurations across local or remote deployments. The system coordinates these settings by applying workspace templates and validating permissions across the agents.

Do I need predefined architecture and memory files before deploying a four-agent OpenClaw setup?

Yes, deploying a four-agent OpenClaw setup requires a predefined architecture, workspace templates, and memory files before initialization. These elements establish the specialized agent roles and enable the validation of permissions and inter-agent networking.

Why does my OpenClaw multi-agent deployment fail during agent-to-agent networking validation?

OpenClaw multi-agent deployment networking validation fails when sandbox settings or tool permissions are misconfigured across agents. Troubleshooting involves checking connectivity, verifying applied workspace templates, and ensuring memory files align with the predefined architecture.

Can I use OpenClaw multi-agent coordination for both local and remote deployments?

Yes, OpenClaw multi-agent coordination applies to both local and remote deployments. It manages agent roles, tool permissions, and sandbox settings across environments by validating permissions and inter-agent networking regardless of deployment location.

How do I troubleshoot connectivity and sandbox issues across specialized agents in a multi-agent setup?

Troubleshooting connectivity and sandbox issues across specialized agents involves validating inter-agent networking and checking tool permissions within the workspace. You verify that memory files and workspace templates are correctly applied to resolve configuration mismatches.