ClawHub

StageWhisper Assistant

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent and non-executable, but it asks an agent to act immediately on live-call tasks that may touch calendars, messages, internal records, and documents without clear approval boundaries.

Install only if you intend the agent to process StageWhisper live-call follow-ups and can restrict its connected tools. Require human review or policy checks before it sends messages, creates calendar events, updates records, or accesses sensitive internal systems.

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 (1)

Vague Triggers

Medium
Confidence
81% confidence
Finding
The skill is described broadly as handling tasks from live calls, without clear scoping to trusted sources, approved task classes, or explicit capability boundaries. That can cause over-invocation or invocation on loosely defined requests, increasing the chance the agent acts on untrusted transcript-derived instructions or performs unintended actions during time-sensitive live-call workflows.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal