ClawHub
Back to skill

Security audit

Harry Tan（Tin Hou Tam）

Security checks across malware telemetry and agentic risk

Overview

This text-only skill does not appear to run code or steal data, but it is designed to help lower plagiarism-check similarity scores on academic papers.

Review this carefully before installing. It has low technical security risk, but its main use is reducing plagiarism-check similarity in academic writing. Use it only for legitimate editing with proper citation, quotation, and original analysis, and do not use it to disguise copied work or bypass institutional academic-integrity rules.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Ssd 4

Medium
Confidence
99% confidence
Finding
The skill gives detailed, procedural guidance for rewriting high-similarity passages into low-similarity wording while preserving the original meaning, including thresholds, tactics, workflow, and quality criteria. In context, this directly enables plagiarism-evasion by helping a user conceal copied material from similarity-detection systems rather than promoting proper citation, quotation, or original authorship.

Ssd 2

Medium
Confidence
97% confidence
Finding
The description uses innocuous academic wording such as '降重改写' and '低相似度原创表达' to package assistance that reduces detectable similarity in thesis text. Even without explicit attack language, the declared purpose is to transform text so it passes plagiarism screening with lower similarity, which meaningfully facilitates academic misconduct.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.