ClawHub

Daily

Security checks across malware telemetry and agentic risk

Overview

This looks like a real Daily video/Membrane integration, but its description misidentifies the data it manages and it can perform sensitive Daily account actions.

Review before installing. Use this only if you intend to connect a Daily.co video account through Membrane, not a CRM system. Require explicit user approval before deleting rooms or recordings, ejecting participants, creating tokens, retrieving recording links, or using raw proxy requests. Consider pinning the Membrane CLI version and revoke the Membrane/Daily connection when no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
The manifest/description says this skill manages CRM-style entities like persons, organizations, deals, and leads, but the body documents a Daily video platform integration with rooms, participants, recordings, and meeting tokens. This mismatch can cause the agent or user to invoke the skill under false assumptions and then perform sensitive actions against the wrong external system, including token creation, participant ejection, or recording access.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger text is broad enough that an agent may select this skill for vague requests about 'Daily data' without clearly scoping the target service or intended action. In the presence of destructive and security-sensitive capabilities like deleting recordings, deleting rooms, ejecting participants, or creating meeting tokens, overbroad routing increases the chance of unintended invocation and unsafe actions.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal