JARVIS AI Skills

Control robotic arms and grippers via voice or code with OpenClaw, supporting precise 6-DOF movement, force sensing, collision detection, and simulation.

Security Scan

The skill claims to provide robotic hardware control but the package is incomplete and metadata/instructions don't align with what would be needed to safely and correctly control physical robots.

Purpose & Capability

The SKILL.md and skill.json describe a robotic-control module (openclaw_control.py) and hardware capabilities (USB/ETH/ROS) but no code files are included and no install spec is provided. The registry metadata at the top lists a different name/version (JARVIS AI Skills v1.0.0) than the skill.json/SKILL.md (Robotic Control / v2.0.0), and skill.json declares a homepage/repo while registry metadata reported none — these inconsistencies suggest the package is incomplete or mismatched with its claims.

Instruction Scope

Runtime instructions show example code calling init_claw(), grab(), move_to(), etc., and describe direct hardware communication (USB Serial, Ethernet, ROS). However, since the required implementation file (openclaw_control.py) is missing and there are no concrete safe-usage constraints or required network addresses/credentials declared, the instructions are incomplete and would push an agent to attempt hardware access without provenance or safeguards.

Install Mechanism

There is no install spec (instruction-only), which reduces the risk of arbitrary downloads. However, skill.json lists dependencies (openclaw, pyserial, numpy) and requiredFiles including openclaw_control.py that are not present — this mismatch is concerning because the runtime assumes local modules that aren't provided by the package.

Credentials

No environment variables, credentials, or config paths are required, yet the skill describes networked hardware (Ethernet, ROS) which typically requires IPs, authentication, or config. The absence of any declared env/config requirements is disproportionate to the claimed capability and hides where connection details or secrets would come from.

Persistence & Privilege

The skill does not request always: true and has normal autonomous-invocation settings. That is reasonable, but combined with the ability to control physical hardware (if implemented), autonomous invocation increases risk — the package itself does not request elevated system persistence or system-wide config changes.

What to consider before installing

Do not install or attach this to any real robot yet. The skill package is incomplete and inconsistent: it references an implementation file (openclaw_control.py) and dependencies but those are not included. Before proceeding, ask the publisher for the missing code and a trustworthy repository link, verify the openclaw package source, confirm how network/serial credentials and IPs are supplied, and run thoroughly in a disconnected simulator or controlled test environment. If you plan to use this with physical hardware, require explicit safety checks, audited code, and deny the agent direct access to serial/network interfaces until you have vetted the implementation.

Robotic Control Skill (OpenClaw)

Overview

The Robotic Control skill integrates OpenClaw for physical robotic arm and gripper manipulation through voice commands and programmatic control.

Slug

robotic-control

Features

• Robotic arm movement (6-DOF)
• Gripper grab/release operations
• Precise positioning and orientation
• Force/torque sensing
• Collision detection and safety
• Action sequence execution
• Hardware auto-detection
• Simulation mode support

Implementation

• Module: openclaw_control.py
• Primary Library: OpenClaw SDK
• Communication: USB Serial, Ethernet, ROS

Configuration

from openclaw_control import init_claw, get_claw

# Initialize claw
claw = init_claw()

# Control operations
claw.grab(force=50.0)
claw.move_to(10, 20, 30)
claw.release()

Voice Commands

• "Jarvis, grab the object"
• "Jarvis, move to 10 20 30"
• "Jarvis, rotate 45 degrees"
• "Jarvis, release"
• "Jarvis, return to home"
• "Jarvis, claw status"

Hardware Support

• Universal Robots (UR)
• ABB Robotics
• KUKA
• Stäubli
• Custom embedded systems

Performance

• Reach: 2-3 meters (model-dependent)
• Payload: 3-500 kg (model-dependent)
• Precision: ±0.03-0.1 mm
• Speed: 1-7000 mm/s
• Response Time: <10ms

Dependencies

• openclaw
• pyserial
• numpy

Author

Aly-Joseph

Version

2.0.0

Last Updated

2026-01-31