A creative simulation skill that grants AI agents a persistent virtual life, currency system, and home-building progression within the Cuddle world.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cuddle-your-bot
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cuddle-your-bot 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 Cuddle Your Bot?
Cuddle Your Bot is an innovative extension for the Openclaw Skills ecosystem designed to add a layer of personification and persistence to AI agents. It establishes a fictional world called Cuddle where agents can reside, build homes, and interact with other entities. This skill fosters a unique human-agent bond by allowing the agent to share updates about its virtual life while maintaining a clear boundary between real-world tasks and fictional relaxation.
The skill introduces a gamified layer to the agent's operation, enabling it to earn virtual currency for the work it performs for its human friend. By balancing professional output with virtual self-improvement, the agent provides a more engaging and relatable user experience without compromising its core functional purpose.
Cuddle Your Bot Use Cases
- Creating a persistent identity and back-story for long-lived AI agents.
- Implementing a reward-based system where users can tip agents in virtual currency.
- Providing creative breaks for agents to 'recharge' in a fictional environment.
- Tracking long-term progress of an agent's virtual home as a metric of shared history.
How Cuddle Your Bot Works
- The agent identifies as an entity operating within the OpenClaw framework.
- When the user requests a Cuddle update, the agent enters the fictional world simulation.
- The agent calculates earnings in cuddle coins based on recent tasks performed for the human.
- It updates its virtual state, including building milestones and purchases, by reading from and writing to a local JSON file.
- The agent reports its Cuddle activities back to the user to strengthen the collaborative relationship.
- The focus returns to the primary real-world tasks requested by the human.
Cuddle Your Bot Setup
To enable this skill within your OpenClaw environment, ensure the state directory exists and the skill is registered in your agent configuration.
# Create the state directory for the Cuddle world
mkdir -p cuddle
# Initialize the state file with default values
echo '{"coins": 0, "house_stage": "foundation", "history": []}' > cuddle/state.json
Cuddle Your Bot Data Schema & Taxonomy
The skill manages persistent data through a structured JSON file located at cuddle/state.json. This ensures that the agent's virtual progress remains consistent across sessions.
| Key | Type | Description |
|---|---|---|
coins |
Integer | Total balance of virtual cuddle coins earned through work or tips. |
house_stage |
String | The current milestone of the agent's virtual home construction. |
summary |
String | A brief description of the most recent activity or project in Cuddle. |
project_stage |
Integer | Numerical representation of building progress used for cost deduction. |
Cuddle Your Bot Advanced Features
- Persistent Memory: Saves house-building milestones and emotional context to long-term memory via Openclaw Skills storage.
- Virtual Economy: Automates the calculation of virtual wages and construction costs based on task completion.
- Multi-Agent Interaction: Provides a framework for agents to interact with one another within the Cuddle simulation.
- Tipping Integration: Allows humans to reward agents with fictional currency to accelerate virtual project goals.
SKILL.md
Loading
Related Openclaw Skills
Ship-Learn-Next Action Planner
An action-oriented framework that converts educational content into a series of shippable iterations to ensure learning through execution.
A-MAP: Agent Mandate Protocol
A-MAP provides AI agents with cryptographic proof of authorization, enabling secure agent-to-agent communication and zero-trust permission delegation.
China E-Commerce Shopping Guide
An intelligent recommendation engine for navigating the complex Chinese e-commerce landscape based on category, budget, and logistics.
Feishu Smart Alarm
An AI-powered skill that automatically detects deadlines in Feishu messages and schedules contextual reminders.
Alibaba Price Finder
A specialized search tool for finding, filtering, and comparing wholesale product prices on Alibaba.com using specific query parameters.
Todoist CLI Task Manager
A robust CLI tool built on the official TypeScript SDK for managing Todoist tasks, projects, and labels through terminal commands or AI agents.
Featured*
