DG-LAB Coyote 3.0 Control Skill for Openclaw

A specialized AI agent skill for controlling DG-LAB Coyote 3.0 pulse hardware via WebSocket and HTTP protocols.

yuuluo
v1.0.0
Mar 14, 2026
0
873
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install dg-lab

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 dg-lab 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 DG-LAB Coyote 3.0 Control Skill?

The dg-lab skill enables AI agents to interact directly with DG-LAB Coyote 3.0 pulse devices. By utilizing a local WebSocket relay, this Openclaw Skills integration allows for precise control over A/B channels, pulse strength adjustment, and the transmission of complex waveform patterns. It acts as a bridge between high-level AI instructions and low-level hardware commands, ensuring a seamless and responsive experience.

Built with a focus on local communication, the skill ensures that control data stays within your local network. It leverages a Python-based client to maintain a persistent connection with the official DG-LAB relay server, providing a stable HTTP API for the AI agent to query device status, generate QR codes for pairing, and execute safety-critical commands like emergency stops.

DG-LAB Coyote 3.0 Control Skill Use Cases

  • Real-time AI-driven control of DG-LAB Coyote 3.0 pulse hardware.
  • Automated pulse strength management based on agent workflows.
  • Dynamic waveform generation and synchronization for hardware interaction.
  • Secure, local-only device pairing for privacy-conscious environments.
  • Emergency safety integration within Openclaw Skills automation.

How DG-LAB Coyote 3.0 Control Skill Works

  1. The AI agent performs an environment check to ensure Python 3, Node.js, and necessary dependencies are present.
  2. A local relay server is cloned from the official DG-LAB Open Source repository and started in the background.
  3. The ws_client.py script initializes a WebSocket connection to the relay and exposes a local HTTP API (default port 8899).
  4. The agent requests a pairing QR code, which the user scans via the DG-LAB mobile app to establish a BLE link to the physical hardware.
  5. Upon successful pairing and safety confirmation, the agent sends commands (strength, waveforms, or presets) via HTTP POST requests to the local client.

DG-LAB Coyote 3.0 Control Skill Setup

Ensure you have Python 3 and Node.js installed, then follow these steps within the Openclaw Skills environment:

# Install Python dependencies
pip install websockets

# Clone official relay server and install Node dependencies
cd ~ && git clone https://github.com/DG-LAB-OPENSOURCE/DG-LAB-OPENSOURCE.gitncd ~/DG-LAB-OPENSOURCE/socket/v2/backend && npm install

# Start the services
npm start &
python scripts/ws_client.py --ws-url ws://localhost:9999 --strength-limit 50 &

DG-LAB Coyote 3.0 Control Skill Data Schema & Taxonomy

The skill interacts via a structured HTTP API and organizes pulse data using specific HEX formats.

Endpoint Method Purpose
/status GET Returns connection and pairing status.
/strength POST Adjusts channel A or B intensity (0-100).
/waveform POST Sends preset or custom HEX pulse data.
/emergency-stop POST Immediately resets all channel strengths to zero.

Custom waveforms are generated as arrays of HEX strings representing frequency and intensity transitions.

DG-LAB Coyote 3.0 Control Skill Advanced Features

  • Multi-channel Support: Independent control of A and B channels for complex setups.
  • Safety Limits: Configurable strength caps via --strength-limit and 100ms rate limiting on commands.
  • Custom Waveform Generator: Python-based waveform.py utility for creating ramp, sine, and pulse patterns.
  • Auto-reconnect: WebSocket client automatically attempts to restore connections every 3 seconds.
  • Extensive Presets: Over 15 built-in patterns including breathing, tide, and heartbeat rhythms.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*