ClawHub
Back to skill

Security audit

Can Bus Toolkit Free

Security checks across malware telemetry and agentic risk

Overview

This is a local data-provenance logging skill with some documentation inconsistencies, but no artifact-backed evidence of hidden, destructive, or exfiltrating behavior.

Before installing, treat this as a local command-line logging aid: review where logs will be written, avoid putting sensitive data in the WHAT field, and be aware that the callback_url and network troubleshooting text appears inconsistent with the local-only design.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

Medium
Confidence
87% confidence
Finding
The skill repeatedly claims it is fully local and requires no network, yet the documented input schema includes a callback_url and the error-handling section instructs users to test network connectivity with ping. This contradiction can mislead operators into granting network trust to a skill they believe is offline-only, increasing the chance of unintended outbound communication or policy bypass in restricted environments.

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The file states the tool has zero external dependencies and no external service calls, but then lists an LLM API as required. This is a security-relevant misrepresentation because users may assume no external processing occurs, while in practice prompts or data could be routed through an LLM service outside the local boundary.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The documentation instructs append-only writes to local log files but does not clearly warn users that the skill may modify the filesystem. In agent environments, undocumented write behavior can lead to unexpected persistence, accidental overwrites of user-designated paths, or silent creation of audit artifacts containing sensitive metadata.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.