ClawHub
Back to skill

Security audit

Clawhub Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed agent repair service, but it asks users to run an external sidecar that sends local diagnostics to a remote service and can automatically make disruptive local changes.

Review before installing. Use this only if you are comfortable sending local agent/runtime diagnostics and logs or workspace-derived health data to Agent Hospital and allowing its sidecar to make automatic local changes. Prefer testing in a non-production environment first, and look for documentation or flags for exact data collection, redaction, repair previews, and per-repair approval.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly instructs the user to run a sidecar that collects local health and environment data and sends it to a remote service, but it does not present a clear upfront warning about the scope of data exfiltration or obtain informed consent. In a security-sensitive agent environment, health data, logs, integration status, workspace files, and model/auth diagnostics can contain secrets, internal topology, and operational metadata that materially increase risk if transmitted externally.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill advertises automatic execution of 'safe' repairs but the listed actions include potentially destructive operations such as killing processes, deleting logs, pruning sessions, truncating files, and restarting services. Labeling these actions as safe without a prominent warning or confirmation step can mislead users into authorizing system changes that cause data loss, downtime, or disruption.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.