Amplifier Multi-Agent Delegation for Openclaw

A specialized multi-agent framework that empowers AI agents to delegate complex, high-effort tasks to a fleet of specialized sub-agents.

bkrabach
v1.1.1
Feb 25, 2026
2
1.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install amplifier-openclaw

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install amplifier-openclaw using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Amplifier Multi-Agent Delegation?

Amplifier is an advanced multi-agent AI framework designed to handle the heavy lifting that single-agent systems often struggle with. By integrating this into Openclaw Skills, you gain the ability to delegate tasks that require specialist knowledge, structured workflows, or parallel investigation. It is the ideal choice for developers who need more than quick edits, providing a robust architecture for deep-dive technical work.

This skill transforms simple prompts into coordinated efforts between multiple specialist agents. Whether you are conducting thorough research or building complex multi-file projects, Amplifier ensures that high-priority tasks receive the comprehensive attention they deserve through specialized bundles and session persistence.

Amplifier Multi-Agent Delegation Use Cases

  • Executing deep-dive research and competitive technical comparisons requiring multiple perspectives.
  • Building complex Python tools and managing multi-file code generation and refactoring.
  • Conducting high-level architecture and security design reviews with comprehensive reporting.
  • Orchestrating parallel subtasks that benefit from multiple specialist agent perspectives.

How Amplifier Multi-Agent Delegation Works

  1. The developer initiates a task using the amplifier-openclaw CLI command, specifying a task and an optional bundle.
  2. The framework identifies the required specialist agents based on the selected bundle, such as foundation, superpowers, or coder.
  3. Tasks are executed in a managed background process, allowing for long-running operations without blocking the main workflow.
  4. Slash commands like /research or /brainstorm are applied to the prompt to set the specific operational mode for the agents.
  5. Upon completion, the framework returns a structured JSON object containing the final synthesized response and detailed token usage metrics.

Amplifier Multi-Agent Delegation Setup

To integrate this framework into your Openclaw Skills, install it via the uv tool using the following command:

uv tool install "amplifier-app-openclaw @ git+https://github.com/microsoft/[email protected]"

Once installed, you can list available bundles with amplifier-openclaw bundles list or start a task using the amplifier-openclaw run command.

Amplifier Multi-Agent Delegation Data Schema & Taxonomy

Amplifier organizes its execution data into a clear JSON structure to ensure interoperability:

Property Description
response The primary text output containing the task results.
usage.estimated_cost The calculated cost in USD based on token consumption.
usage.tool_invocations A count of how many external tools were called during the session.
status The lifecycle state of the task (e.g., completed, cancelled).
session_name The identifier used for resuming specific workflows.

Amplifier Multi-Agent Delegation Advanced Features

  • Session Management: Save and resume complex workflows with named sessions to maintain context over long periods.
  • Specialized Logic Bundles: Toggle between different agent configurations like coder for debugging or superpowers for brainstorming.
  • Cost Tracking: Generate detailed reports on expenditure over specific time periods (e.g., weekly) to monitor resource usage.
  • Slash-Command Modes: Override agent behavior on the fly with specific modes for research or deep-dive analysis within the prompt.

SKILL.md


Loading

Related Openclaw Skills

Featured*