Conversion Copywriting Engine

• Outcome: Pass
• Latest audit: Sep 10, 2026 · 11:34 PM UTC
• Version: 1.0.0

Overview

This is a Markdown-only copywriting skill with broad marketing guidance, not hidden code or unsafe install behavior.

Before installing, treat this as a marketing persuasion and copywriting skill. Do not provide private support tickets, call transcripts, customer records, regulated data, or platform content unless you have authorization and have redacted sensitive details. If your agent auto-routes skills, consider requiring explicit invocation for this skill because its example commands are broad.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The README encourages mining customer language from reviews, tickets, and forums but does not warn about privacy, confidentiality, terms-of-service, or handling of personal/sensitive data. In practice, users may paste support tickets, private forum content, or regulated customer data into the workflow, creating a risk of inappropriate collection, disclosure, or policy noncompliance.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The quick-start examples are generic, natural-language requests like 'Write a landing page' and 'Score this copy,' which are broad enough to overlap with ordinary user prompts. In systems that auto-route or invoke skills based on prompt similarity, this can cause unintended invocation of the skill in contexts where the user did not explicitly request it, potentially exposing business content to the skill unnecessarily or producing unexpected behavior.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The description is very broad and can be triggered by many ordinary writing or review requests, which increases the chance of unintended skill activation. Over-broad routing can cause this persuasive copywriting skill to be invoked in contexts where a more appropriate or safer skill should handle the task, expanding access to prompt content and user data unnecessarily.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The natural-language commands are generic imperative phrases such as 'Score this copy' and 'Rewrite this copy to convert better,' which lack scope guards and may match commonplace assistant requests. This makes accidental or overly eager invocation more likely, potentially hijacking benign writing workflows and applying persuasion-oriented behavior outside intended marketing contexts.

Static analysis

No suspicious patterns detected.