ClawHub
Back to skill

Security audit

Owner Briefing

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed daily briefing skill that collects calendar, email, and task information and sends a summary, but it is privacy-sensitive and should be configured carefully.

Install or schedule this only if you are comfortable sending meeting titles, email senders/subjects, and task names through the selected delivery channel. Before enabling cron, verify the Google account, monday.com board ID, token scope, OWNER_PHONE or email recipient, timezone, and whether WhatsApp is appropriate for your organization’s data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill reads a local monday.com API token from a credential file and uses it to access third-party task data. Although that access supports the stated briefing purpose, it still expands the skill's privilege surface to local secrets and external SaaS data, and the manifest does not clearly disclose this sensitive credential usage to the user.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The manifest description includes broad triggers such as 'when asked for a summary' or scheduled use, which can cause the skill to run in ordinary contexts without a clear, narrow invocation boundary. Because the skill accesses private calendar, email, and tasks and can transmit a compiled briefing, over-broad invocation increases the chance of unintended privacy-sensitive execution.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill description does not clearly warn that it reads private calendar, Gmail, and monday.com task data and then sends the compiled output over WhatsApp or email. This lack of transparent disclosure undermines informed consent and can lead to sensitive personal or business information being transmitted through channels the user did not fully expect.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.