ClawHub

自媒体运营全能助手

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Chinese social media operations guide with no code, installs, credentials, or system access requested.

Safe to install as a marketing guidance skill. Treat the custom-service contact as optional third-party promotion, and avoid sharing sensitive business, customer, or account data outside the agent unless you intentionally choose to contact the author.

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
86% confidence
Finding
The skill is user-invocable and broadly framed as a general-purpose social media assistant, so it may activate on many ordinary requests that only loosely relate to its intended scope. Over-broad activation can cause inappropriate routing, unexpected behavior, or overshadow more suitable skills, which is a real quality and safety issue even if it does not directly enable code execution or data exfiltration.

Natural-Language Policy Violations

Medium
Confidence
78% confidence
Finding
The metadata and content are entirely Chinese-oriented, with no indication that language should follow user preference or that the skill is restricted to Chinese-speaking users. This can cause the skill to respond in an unexpected language, reducing usability and potentially leading to misunderstandings in generated marketing guidance.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal