Create New Openclaw Instance on GCP VM

Security checks across malware telemetry and agentic risk

Overview

This is a coherent cloud setup guide, but it asks you to run powerful deployment commands, install software from the internet, and configure persistent credentials on a new VM.

Before using this skill, review each command, use a dedicated GCP project if possible, confirm expected cloud costs, inspect any missing quick-setup script before running it, and be comfortable with installing persistent OpenClaw and Tailscale services on the VM.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

Installing or following this skill can create cloud resources and place provider API keys on the VM.

Why it was flagged

The skill requires cloud account context and API credentials. These are expected for the stated deployment, but they are sensitive and the registry metadata lists no required credentials or environment variables.

Skill content
export OPENCLAW_PROJECT_ID="your-gcp-project" ... export ANTHROPIC_TOKEN="sk-ant-oat01-..." ... export BRAVE_API_KEY="..."
Recommendation

Use a dedicated GCP project and scoped API keys where possible, confirm expected costs, and rotate credentials if they are exposed.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

The VM will execute code fetched from the internet, and the exact OpenClaw package version may change over time.

Why it was flagged

The manual setup installs software from remote scripts and installs the latest OpenClaw package. This is common for setup automation, but it relies on external sources at execution time.

Skill content
curl -fsSL https://tailscale.com/install.sh | sh ... curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.1/install.sh | bash ... npm install -g openclaw@latest
Recommendation

Review the install sources, pin package versions where practical, and run the commands only in a VM/project you are comfortable modifying.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

A user might look for or run a setup script that was not included in the reviewed artifacts.

Why it was flagged

The Quick Start references a helper script, but the provided manifest contains only SKILL.md. This is an incomplete review context rather than evidence of malicious behavior.

Skill content
chmod +x openclaw-quick-setup.sh
./openclaw-quick-setup.sh
Recommendation

Do not run any separately obtained quick-setup script unless you inspect it first or obtain it from a trusted source.

#
ASI02: Tool Misuse and Exploitation
Medium
What this means

If copied blindly, the commands may accept a wrong SSH host key or skip prompts that would otherwise explain risk before configuration.

Why it was flagged

The setup disables SSH host key checking for the first connection and uses a non-interactive risk-accepting onboarding command. These choices streamline setup but reduce interactive safety checks.

Skill content
ssh -o StrictHostKeyChecking=no "${OPENCLAW_USERNAME}@${IP}" ... openclaw onboard --non-interactive --accept-risk
Recommendation

Verify the VM IP and SSH host key when possible, and understand what OpenClaw onboarding will install before using '--accept-risk'.

#
ASI10: Rogue Agents
Medium
What this means

The VM will continue running OpenClaw-related services and may remain reachable over Tailscale until stopped or removed.

Why it was flagged

The deployment intentionally creates persistent background services for the OpenClaw gateway and Tailscale serving. This is aligned with hosting OpenClaw, but it keeps running after setup.

Skill content
openclaw onboard ... --gateway-bind loopback --install-daemon ... sudo tailscale serve --bg 18789
Recommendation

Confirm you want a persistent gateway, monitor the VM and Tailscale serve status, and document how to stop or uninstall the services.