ClawHub
Back to skill

Security audit

Smartclaws Bridge Agent

Security checks across malware telemetry and agentic risk

Overview

This appears to be a disclosed telemetry bridge skill, but users should invoke its hardware-read and on-chain publish actions deliberately.

Install only if you intend the agent to interact with the relevant local sensor/device setup and publish telemetry on-chain. Before using it, confirm what data will be published, what account or wallet pays for transactions, and require explicit confirmation for real publish actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The skill advertises broad natural-language triggers such as being asked to 'read the sensor and publish' or 'run a telemetry/bridge cycle,' which can cause accidental or overly eager invocation in contexts where the user did not intend a real hardware read or on-chain publish. Because this skill can interact with local device interfaces and publish data on-chain, unintended activation can lead to unauthorized actions, unnecessary transactions, or disclosure of operational telemetry.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.