ClawHub
Back to skill

Security audit

Heleni Meetings

Security checks across malware telemetry and agentic risk

Overview

This meetings skill is useful and purpose-aligned, but it persistently saves full raw meeting notes by default and can modify calendars without consistently requiring explicit approval.

Review before installing if your meetings may include confidential, legal, HR, customer, or credential-related content. Use it only with an approved calendar account, require confirmation before creating or deleting calendar events, and avoid saving raw transcripts unless users explicitly want that local retention.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill explicitly instructs saving meeting notes to disk and preserving the full raw notes/transcript, which can include sensitive business discussions, personal data, credentials, or regulated information. Because there is no consent step, sensitivity warning, retention guidance, or redaction requirement, the skill increases the risk of unintended local data persistence and later exposure through workspace access, backups, or indexing.

Missing User Warnings

Low
Confidence
85% confidence
Finding
The skill directs the agent to write extracted action items into `todo.md`, which is a workspace modification, without clearly warning the user that a file will be created or changed. This can lead to unexpected state changes, accidental persistence of sensitive tasks, and downstream confusion if users did not intend their workspace files to be modified.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.