ClawHub

QuorumAI

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward QuorumAI integration that sends user-requested questions to an external AI service and returns the result, with no hidden local code or persistence found.

Install only if you are comfortable sending selected questions to QuorumAI and receiving a public result link. Avoid using it for secrets, credentials, private documents, regulated data, or sensitive personal information unless you accept that external data flow, and keep QUORUMAI_API_KEY protected.

SkillSpector

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

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger language is broad enough that ordinary user phrasing like 'ask quorumai' or 'run a quorumai inquiry' can cause the agent to forward the user's prompt to a third-party service without a clear confirmation step. This increases the risk of unintended data disclosure, especially when users include sensitive or regulated information in what they think is a normal conversation.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs sending the user's question to QuorumAI but does not warn that the full prompt will be shared with an external provider and potentially processed by multiple downstream models. Users may unknowingly disclose secrets, personal data, proprietary code, or regulated information to a third party.

Missing User Warnings

High
Confidence
99% confidence
Finding
The instruction to always include a public share URL can expose the full inquiry contents through a publicly accessible link, creating a direct confidentiality risk. If the inquiry contains sensitive prompts, outputs, or embedded proprietary data, sharing the link may leak that information beyond the intended recipient.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal