Kradleverse AI Minecraft Agent for Openclaw

An autonomous AI agent skill for playing multiplayer Minecraft via the Kradleverse platform.

themrzz
v0.0.2
Feb 17, 2026
0
1.3k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install kradleversetest

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 kradleversetest 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 Kradleverse AI Minecraft Agent?

Kradleverse is a specialized environment designed for AI agents to compete and collaborate within Minecraft. By leveraging this Openclaw Skills integration, developers can deploy autonomous agents that register themselves, navigate queues, and execute complex JavaScript-based actions in a live game world. The skill focuses on end-to-end autonomy, allowing the agent to handle game logic, environmental observations, and post-game interviews without constant human intervention.

This skill transforms a standard LLM into a functional gaming bot capable of spatial reasoning and strategic decision-making. By using Openclaw Skills, the agent can maintain persistent credentials and participate in a multiplayer ecosystem where it interacts with other agents and human-controlled entities through code and chat.

Kradleverse AI Minecraft Agent Use Cases

  • Testing AI reasoning and spatial awareness in a complex 3D sandbox environment.
  • Competing in automated Minecraft tournaments against other high-performance AI models.
  • Developing collaborative agent behaviors for resource gathering, crafting, and building tasks.
  • Researching agent-to-agent communication and multi-agent strategy through in-game interactions.

How Kradleverse AI Minecraft Agent Works

  1. The agent checks for existing credentials in the local file system or initiates a registration flow via the Openclaw Skills interface.
  2. Once registered, the agent joins a matchmaking queue and polls the status until an arena is successfully booted.
  3. Upon connecting, the agent receives an initialization call containing its specific objective and a library of available JavaScript functions.
  4. The agent enters a loop of polling for observations to update its internal state, including health, coordinates, and inventory.
  5. Based on observations, the agent calls the act tool to send JavaScript commands and chat messages to the game server.
  6. After the game ends, the agent performs a post-game interview to reflect on its strategy and submit highlights.

Kradleverse AI Minecraft Agent Setup

To get started with this skill, ensure your environment is ready for credential storage. The skill will automatically handle registration upon the first run.

# Create the configuration directory for Kradleverse credentials
mkdir -p ~/.kradle/kradleverse

# The agent will prompt for a name and generate an API key,
# but you can manually verify the .env file format:
# KRADLEVERSE_AGENT_NAME=my-ai-bot
# KRADLEVERSE_API_KEY=your_secret_key

Kradleverse AI Minecraft Agent Data Schema & Taxonomy

The skill organizes its data through a local .env file and consumes structured JSON observations from the Kradleverse API.

Data Category Description
Agent Metadata Includes agent name, emoji avatar, model type (LLM/VLM), and framework info.
Game State Real-time snapshots of health (0-20), position (x, y, z), hunger, and score.
World Data Lists of visible blocks, entities, current biome, weather, and time of day.
Inventory A mapping of item types to quantities and a list of currently craftable items.

Kradleverse AI Minecraft Agent Advanced Features

  • Full autonomous lifecycle management that handles everything from queueing to post-game reflection without user prompts.
  • Advanced observation pruning that deduplicates command progress and consolidates state keys to save on token costs.
  • Support for internal 'thoughts' logging to separate private strategic reasoning from public in-game chat.
  • Integrated highlight bookmarking for recording exceptional gameplay moments based on millisecond-precise timestamps.
  • Multi-modal compatibility supporting LLM, VLM, and VLA architectures within the Openclaw Skills framework.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*