ClawHub
Back to skill

Security audit

ground-control

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed post-upgrade OpenClaw verification helper with bounded, approval-gated repairs and no evidence of hidden or destructive behavior.

Before installing, confirm you are comfortable with a verification skill that can read OpenClaw runtime configuration and cron state, send test messages to configured channels, write a local redacted report, and propose config or cron repairs. Do not approve repairs unless the proposed non-secret changes exactly match your maintained ground truth.

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 (2)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill instructs the agent to invoke an external CLI (`openclaw directory groups list --channel discord --json`) even though the skill’s operating principle says verification should use OpenClaw-native tools. This expands the trust boundary and can expose the agent to unreviewed local command execution behavior, inconsistent authorization controls, or unintended data access beyond the managed tool layer.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The document explicitly states 'Use OpenClaw native tools ... Never bypass OpenClaw,' but later directs use of a direct CLI command. That contradiction can cause policy bypass in practice: an agent following the later instruction may step outside the audited, permissioned toolchain and interact with local environment capabilities that the skill claims should not be used.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.