agent-collab

Coordinate multi-agent collaboration across departments within OpenClaw.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/jakechen1/echo-research-framework --skill agent-collab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-collab
Source: https://github.com/jakechen1/echo-research-framework/tree/main/skills-available/agent-collab
Command: npx skills add https://github.com/jakechen1/echo-research-framework --skill agent-collab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenClaw often suffers from fragmented multi-agent coordination across departments. This Skill defines three interaction modes—Dispatch, Collaborate, and Direct Chat—and provides an org hierarchy, HANDOFF mechanism, and scoped memory to streamline cross-agent workflows.

Core Features & Use Cases

  • Dispatch mode for one-shot tasks with automatic result delivery.
  • Collaborate mode for persistent, cross-department sessions with context sharing.
  • Direct Chat mode for user-initiated, department-head discussions.
  • HANDOFF push mechanism for asynchronous task handovers between agents.
  • Scoped memory and org hierarchy to maintain context and governance.

Quick Start

Outline your task and use spawn with an appropriate mode to begin dispatching to the target agent.

Frequently Asked Questions about agent-collab

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

FAQPage Schema
How do I coordinate multi-agent tasks across different departments in OpenClaw?

Multi-agent tasks in OpenClaw are coordinated through defined interaction modes like Dispatch, Collaborate, and Direct Chat. These modes utilize an org hierarchy and HANDOFF mechanisms to streamline cross-department workflows and ensure asynchronous task handovers between agents.

What is the best way to handle persistent cross-department sessions with multiple agents?

Persistent cross-department sessions are best handled using Collaborate mode. This mode maintains continuous context sharing across participating agents while leveraging scoped memory isolation to preserve governance boundaries throughout the workflow.

How does the HANDOFF mechanism work for asynchronous task handovers?

The HANDOFF mechanism works as a push process for asynchronous task handovers between agents. It allows one agent to transfer a task payload to another within the governance structure, ensuring the receiving agent picks up the context without direct synchronous coupling.

When should I use Dispatch mode versus Direct Chat mode for agent workflows?

Use Dispatch mode for one-shot tasks that require automatic result delivery, and use Direct Chat mode for user-initiated discussions with department heads. Dispatch handles automated single-shot execution, while Direct Chat enables interactive, user-driven conversations.

Can I isolate agent memory during cross-department collaboration?

Yes, memory isolation is supported during cross-department collaboration. Scoped memory boundaries are applied to maintain context governance, ensuring that agents only access the information designated for their specific departmental role.

How do I start dispatching a single-shot task to a target agent?

To dispatch a single-shot task, outline your task requirements and use the spawn command with an appropriate mode. This initiates the dispatch process to the target agent, which will execute the task and deliver the results automatically.