ClawHub

Phone Call - CALL-E

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed real phone-calling skill, but it gives the agent weakly scoped authority to place calls and display sensitive call details.

Review before installing. Use this only if you trust CALL-E, are comfortable authorizing a CLI to place real outbound calls, and understand that call details and transcripts may be shown in chat. Double-check the number and purpose before any call, and prefer an installed or pinned official CLI over a repository-local CLI unless you trust that workspace code.

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 (2)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The invocation text is broad enough to trigger on common phrases like 'call a business' or 'follow up by phone,' which can cause the agent to select a real-world telephony skill in situations where the user may only want advice, drafting, or simulation. Because this skill can place actual outbound calls, over-broad routing materially increases the risk of unintended external actions.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill provides execution instructions for placing real outbound phone calls without an explicit upfront warning about privacy, consent, and the real-world impact of contacting external parties. In a telephony skill, this omission increases the risk of users initiating calls without understanding disclosure, legal, or harassment implications, especially because the commands move directly from planning to execution.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal