ClawHub
Back to skill

Security audit

PingBell

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for PingBell access, but it presents itself as read/search oriented while also enabling an untagged state-changing ring action through connected account credentials.

Review before installing. This skill can list PingBell sources and can also ring a source using your connected PingBell account. Only use it where ringing a source is acceptable, and require explicit confirmation of the target and expected effect before any ring_source call.

SkillSpector

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

Medium
Confidence
97% confidence
Finding
The manifest description says the skill is for "searching and reading data," but the documented actions include `ring_source`, which changes external state by causing a PingBell source to ring and update subscribed devices. This mismatch can mislead an agent or user into treating the skill as read-only, reducing scrutiny and increasing the chance of unintended real-world actions.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The phrase "Use this skill for ANY PingBell request" is overly broad and encourages indiscriminate routing of all PingBell-related tasks through a skill that includes a write action. In an agent setting, broad trigger language increases the likelihood that the skill is invoked in contexts where a narrower read-only skill or direct user confirmation would be safer, especially because OOMOL injects credentials server-side and enables real actions immediately.

Static analysis

No suspicious patterns detected.