ClawHub

上海话（沪语）

Security checks across malware telemetry and agentic risk

Overview

This is a text-only Shanghai dialect conversation skill whose main risk is confusing language-mode behavior, not security-sensitive access.

Install only if you want the assistant to speak Shanghai dialect by default. Be aware it may switch into dialect mode on weak cues; use the documented normal-language mode if that becomes inconvenient. Security risk is low because it is instruction-only and does not use code, credentials, local data, or network access.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
95% confidence
Finding
The trigger detection rules are broad enough to activate on common words such as '啥' or '蛮', as well as short abbreviations like 'shh', which can appear in unrelated conversations. This can cause unintended mode switching or dialect responses without clear user intent, reducing reliability and potentially conflicting with user expectations or platform language behavior.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
Setting Shanghainese as the default response mode without explicit opt-in can override the user's expected language preference from the start of the interaction. While not a classic security flaw, it is a genuine policy and safety concern because it can impair comprehension, create confusing outputs, and cause the assistant to behave in a way the user did not request.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal