ClawHub
Back to skill

Security audit

Cart Bundle Aov Upsell

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only ecommerce guidance skill for cart bundles and upsells, with no code execution, credentials, persistence, or privileged access.

Installers should treat this as a low-risk sales-copy and merchandising helper. Review its broad trigger wording if you need strict routing, especially for general ecommerce marketing tasks, but it does not show security-sensitive behavior.

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 (2)

Vague Triggers

Medium
Confidence
97% confidence
Finding
The description includes very broad activation cues such as 'make people spend more per order' and several loosely related ecommerce intents, which can cause the skill to trigger outside narrowly scoped cart-bundling scenarios. This increases the chance of inappropriate invocation, context confusion, and the model producing bundling guidance when a different skill or safer workflow should handle the request.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The eval expectations for the SEO meta-description case explicitly say the skill should not force the bundle table, which implies the triggering logic may otherwise be broad enough to engage on unrelated ecommerce requests. If activation criteria are underspecified, the skill can inappropriately hijack non-cart tasks, causing incorrect outputs, workflow interference, and possible chaining into downstream automation that assumes the task was correctly classified.

Static analysis

No suspicious patterns detected.