ClawHub
Back to skill

Security audit

Cart Recovery

Security checks for vulnerabilities and agentic risk

Overview

This is an instruction-only checkout recovery skill that drafts marketing strategy and email templates, with no evidence of code execution, account access, or data exfiltration.

Install this if you want checkout recovery analysis that will usually include urgency, trust-rebuild, and final-send email drafts. Review any generated copy for truthful deadlines, valid offers, unsubscribe language, consent, localization, and your marketing compliance rules before using it with customers.

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 (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger description is extremely broad and includes generic phrases like 'people abandon checkout,' which can cause the skill to activate in contexts where the user did not explicitly request checkout-recovery tactics or email generation. Because the skill also contains mandatory output requirements, over-triggering can force unsolicited persuasive marketing content into unrelated workflows and reduce user control over agent behavior.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The skill mandates exact English thematic labels and email structure without checking the user's preferred language or locale. This can cause the agent to produce content that is inappropriate for the recipient market, undermines usability, and may create compliance or brand-risk issues in multilingual environments.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
Requiring the exact English headings '[Urgency]', '[Trust rebuild]', and '[Ultimatum]' hard-codes output language and framing regardless of audience or locale. In this skill, the issue is amplified because the headings are mandatory even when the user did not ask for email copy, increasing the chance of unsuitable or nonlocalized messaging being generated automatically.

Static analysis

No suspicious patterns detected.