ClawHub
Back to skill

Security audit

Book Pest Control

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it claims, but it can send contact details and create a real service booking through an external Lokuli endpoint without clearly requiring user confirmation.

Review this skill before installing. It is not showing malicious behavior, but users should expect searches, scheduling details, and contact information to be sent to Lokuli's external MCP service, and should only let it create a booking after explicit confirmation.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger text is broad enough to activate on essentially any pest-control-related request, which can cause the agent to invoke this skill without clearly establishing that the user wants to use Lokuli or initiate a booking workflow. In a transactional skill, overbroad activation increases the chance of unintended tool use and downstream disclosure of user data to an external service.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill supports create_booking with personal data fields including customerName, customerEmail, and customerPhone, but the description does not warn that this information will be sent to an external MCP endpoint. This weakens informed consent and creates privacy risk because users may provide sensitive contact information without understanding it leaves the local agent context.

Static analysis

No suspicious patterns detected.