AutoGen Skill for Openclaw for Openclaw

A sophisticated multi-agent orchestration framework designed to automate complex game design and development workflows through autonomous AI collaboration.

kaiyuelv
v1.0.0
Apr 11, 2026
0
813
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install autogen-skill

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 autogen-skill 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 AutoGen Skill for Openclaw?

The AutoGen skill brings the power of Microsoft's AutoGen framework into the Openclaw ecosystem. It enables the creation of multiple AI agents that can converse with one another to solve complex tasks, specifically optimized for game development scenarios. By integrating this into Openclaw Skills, developers can build systems where different agents take on specialized roles, such as game designers, coders, or testers, working together to produce high-quality outputs with minimal human intervention.

This skill provides the necessary bridge to implement autonomous agentic workflows within your workspace. It handles the nuances of agent communication, LLM configuration, and state management, making it an essential component for any developer looking to push the boundaries of AI-driven creativity and automated project management.

AutoGen Skill for Openclaw Use Cases

  • Automating the creation of multi-chapter RPG storylines and world-building documentation.
  • Simulating collaborative game design sessions between specialized AI agents.
  • Orchestrating automated code generation and debugging cycles for complex game mechanics.
  • Creating self-improving feedback loops for game balancing and narrative design.

How AutoGen Skill for Openclaw Works

  1. Initialize specialized agents by defining their roles and assigning specific LLM configurations.
  2. Configure a UserProxyAgent to act as the interface between the human developer and the autonomous agent swarm.
  3. Initiate a chat session with a specific objective or prompt, such as designing a game level.
  4. Agents engage in a multi-turn dialogue, exchanging information and refining the output based on their assigned personas.
  5. The workflow concludes when the agents reach a consensus or complete the defined task, outputting the final design or code.

AutoGen Skill for Openclaw Setup

The AutoGen skill is pre-installed in the Openclaw environment. To use or update it in your workspace, follow these steps:

# Navigate to the skill directory
cd /workspace/skills/gamedev-tools/autogen/

# Ensure the core library is installed
pip install pyautogen

AutoGen Skill for Openclaw Data Schema & Taxonomy

The AutoGen skill utilizes a structured approach to agent communication and task management. Below is the primary metadata and file structure:

Component Type Description
AssistantAgent Object Defines the AI persona, instructions, and LLM backend settings.
UserProxyAgent Object Manages execution environments and human-in-the-loop triggers.
llm_config Dictionary Stores API keys, model versions, and temperature settings.
Chat History Log Maintains the conversational context for multi-agent persistence.

AutoGen Skill for Openclaw Advanced Features

  • Support for complex conversation patterns including group chats and hierarchical agent structures.
  • Customizable human-input modes (NEVER, ALWAYS, TERMINATE) for varying levels of oversight.
  • Seamless integration with external tools and APIs through function calling within the agent workflow.
  • Capability to handle multi-modal tasks by combining different LLM backends within a single Openclaw Skills session.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*