ClawHub
Back to skill

Security audit

Ab Test Setup

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward A/B testing guide with only a minor risk of activating too broadly in comparison-related conversations.

Installers should expect this skill to help with experiment design and may want to narrow when it activates if they have other analytics or CRO skills installed. The optional local context-file read is disclosed and relevant, but users should keep sensitive unrelated material out of those context files.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description includes broad activation language such as "When the user wants to plan, design, or implement an A/B test or experiment" and especially "Use this whenever someone is comparing two approaches and wants to measure which performs better." That scope can overlap with many general decision-making conversations, making it unclear when this skill should activate versus more specialized alternatives.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.