ClawHub
Back to skill

Security audit

Beebole

Security checks across malware telemetry and agentic risk

Overview

The skill is a Beebole connector, but it presents itself as mainly for reading while exposing a generic GraphQL action that can also mutate connected account data.

Review this skill before installing if you want a read-only Beebole helper. It may be useful for Beebole automation, but users should require explicit confirmation for every GraphQL mutation and verify the exact payload and affected records before allowing it to run.

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
97% confidence
Finding
The skill metadata and description claim it should be used for 'searching and reading data,' but the only exposed action is `execute_graphql`, which explicitly allows queries or mutations. That creates a scope-deception issue: an agent may invoke this skill under the assumption it is read-only while actually being capable of modifying or deleting Beebole data if prompted or if a mutation is constructed.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The description says to use this skill for 'ANY Beebole request' and to prefer it over direct API access, which is overly broad and lacks task-level scope limits. Because the skill exposes a generic GraphQL executor, this broad routing language increases the chance an agent will send high-risk or unintended operations through it without sufficient narrowing, review, or policy gating.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.