ClawHub
Back to skill

Security audit

Code Helper

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a coherent competitive-programming tutor, but its broad auto-activation plus under-specified Feishu team-document export creates review-worthy risk.

Install only if you want this skill to activate broadly for programming-help requests. Before using the Feishu export/archive feature, confirm exactly what content will be sent, which Feishu workspace or team document will receive it, and whether you can preview or cancel the write first.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

High
Confidence
95% confidence
Finding
The trigger list is overly broad and includes common terms like “代码”, “题目”, “思路”, and “算法”, which can cause the skill to activate in many unrelated conversations. Over-broad auto-activation increases the chance of unintended instruction takeover, unexpected behavior, and accidental invocation of any downstream capabilities mentioned in the skill.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill advertises one-click export to Feishu documents but does not clearly disclose that user content may be written to an external service. This can lead to inadvertent sharing of problem statements, code, notes, or other potentially sensitive material without informed user consent.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The “飞书归档命令” implies an external write action, but the skill provides no privacy warning, no consent checkpoint, and no description of system effects. Any capability that persists user-generated content to third-party systems should be treated as sensitive because it may expose data or create unintended records.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.