Clawbus Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a creator-workflow installer, but it can download, overwrite, and immediately activate remote skill instructions with broad triggers.

Install only if you trust ClawBus as a source for downstream skills. Before allowing activation, review the downloaded files, confirm the exact slug and destination account, and avoid letting broad creator-workflow requests automatically refresh or overwrite local skills.

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 (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The top-level description says to use this skill whenever the user asks about ClawBus, installing a skill, or using a ClawBus slug, which is broad enough to trigger remote skill installation and activation from normal conversational mentions. Because the activation flow later instructs the agent to fetch, write, read, and follow downloaded files, overbroad invocation increases the chance of unintended execution of untrusted remote instructions.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The 'When To Use' section includes ambiguous triggers like mentions of creator publishing, YouTube APIs, Instagram publishing, or TikTok publishing, plus a bare 'use <slug>' pattern. In this skill's context, those triggers are more dangerous because they lead directly into a supply-chain style flow that downloads arbitrary skill files from a remote endpoint, overwrites local skill directories, and follows the downloaded SKILL.md instructions in the same session.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal