ClawHub
Back to skill

Security audit

OpenClaw Studio

Security checks across malware telemetry and agentic risk

Overview

This appears to be a local OpenClaw dashboard, but its installer creates persistent background services whose runtime code is not included for review.

Review before installing. Prefer ./run_monitor.sh or ./start_bg.sh for local use, and do not run install_launchd.sh unless you specifically want persistent macOS background services and have reviewed the missing runtime files. If persistence was enabled, use uninstall_launchd.sh to remove the LaunchAgents.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The script installs three persistent LaunchAgents and immediately enables and starts them, which goes beyond a simple one-time local dashboard launcher. Persistence itself is not always malicious, but here it materially increases the attack surface and establishes long-running background execution without visible user consent or tight scope limitation.

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The auto-heal and watchdog agents introduce self-recovery and periodic supervision behavior that is not clearly justified by a dashboard-presenter skill. Such components can frustrate shutdown, re-establish execution after failure, and make unwanted behavior harder for a user to detect or remove.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The script writes LaunchAgent plist files into the user's persistence directory and enables them without any user-facing warning, prompt, or confirmation. Silent persistence is dangerous because users may believe they are only launching a local dashboard, while the skill actually installs background services that survive across sessions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.