Security audit

Email Sequence

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only email marketing guidance skill with some privacy-compliance caveats, but no hidden execution, persistence, or destructive behavior.

Before using this skill for real campaigns, make sure any personalization or behavior-based triggers rely on data you are allowed to use, avoid sensitive attributes, and comply with consent, unsubscribe, and regional privacy rules. The skill is suitable as strategic writing guidance; treat implementation through email platforms as a separate step requiring normal review and approvals.

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 (5)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The manifest description says to use the skill when the user wants to create or optimize an email sequence, and also includes broad phrases like "what emails should I send" and "Use this for any multi-email automated flow." Those phrases are expansive enough to overlap with many general marketing requests, without clear boundaries or negative examples beyond two adjacent skills.

Missing User Warnings

Medium

Confidence: 81% confidence
Finding: The skill guides the agent to reason about audience context, triggers, beliefs, and relationship state without any privacy guardrails around use of customer or behavioral data. In practice, this can normalize collecting or processing sensitive segmentation attributes without user warnings, minimization, or consent checks, increasing the risk of privacy violations or inappropriate profiling.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: This JSON manifest/evals file includes expectations that the skill 'Should trigger on casual phrasing' and asserts 'Triggers on casual phrasing.' That activation guidance is ambiguous and broad enough to overlap with everyday speech, increasing the risk of unintended invocation.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: This guidance encourages behavior-based and personalized outreach using product-usage signals (for example, referencing what a user did or did not do) without any caution about consent, transparency, data minimization, or jurisdiction-specific privacy requirements. In a marketing automation skill, that omission can lead implementers to deploy intrusive or non-compliant messaging that exposes sensitive behavioral data, increases regulatory risk, and erodes user trust.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: This markdown file gives operational guidance for using merge fields, dynamic content, triggered emails, and segmentation by behavior/profile, which can affect user privacy and data handling. Under SQP-2 for markdown files, the description should warn that these practices rely on user data and should be used transparently and in compliance with consent/privacy requirements.

Static analysis

No suspicious patterns detected.