Security audit

Audience Builder

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent ad-audience planning guide, but it repeatedly directs use and third-party upload of customer identifiers and value data without adequate privacy or compliance guardrails.

Review this skill before installing if your campaigns use real customer data. Only use the workflows where you have documented consent or another lawful basis, exclude opted-out and sensitive-category data, upload the minimum identifiers needed, use approved hashed upload methods, and confirm each platform/vendor contract, retention, deletion, and regional privacy requirement.

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
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (7)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill explicitly instructs users to export customer files, segment them, and upload Customer Match/custom audience lists across ad platforms without any warning to verify consent, lawful basis, platform policy compliance, or minimization of personal data. In an ad-tech context, this can lead to privacy violations, improper sharing of personal data with third parties, and regulatory exposure if operators follow the workflow blindly.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The guide explicitly instructs users to upload directly identifying customer data such as email, phone, names, and location fields to ad platforms, but it provides no privacy, consent, minimization, retention, or jurisdictional compliance guidance. In an audience-building skill, this omission is material because it normalizes transfer of customer PII to third parties and can lead operators to violate privacy laws, platform terms, or internal data-handling policies.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The document recommends uploading lifetime value data to Meta for value-based lookalikes without warning that customer-value attributes are sensitive commercial profiling data and may require additional disclosure, consent, and governance. In this skill's marketing context, the feature is operationally relevant, which makes the omission more dangerous because users are likely to follow it in production and share enriched customer profiles with a third party.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The template explicitly guides use of customer-file uploads and enumerates identity fields like email, phone, and MAID, but provides no guardrails about consent, lawful basis, minimization, retention, or platform policy requirements. In an advertising-audience skill, this omission can lead users to process or upload sensitive personal data improperly, creating privacy, compliance, and account-enforcement risk even if the content is not overtly malicious.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The implementation timeline operationalizes customer file segmentation and upload as a routine step without any warning about privacy review, consent validation, or secure handling. Because this step is action-oriented, it increases the chance that an operator will upload personal data directly into ad platforms without verifying authorization or data-protection requirements.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The document instructs users to upload customer contact data and mobile/device identifiers for audience creation without any accompanying privacy, consent, retention, or lawful-basis guidance. In an ad-targeting skill, this omission can lead users to process regulated personal data in ways that violate platform policies or privacy laws such as GDPR/CCPA, creating compliance and data-sharing risk.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The cross-platform sync section recommends moving audience data through CDPs, APIs, Zapier/Make, and identity platforms but does not warn that these transfers expand third-party exposure and may require user notice, consent, vendor agreements, and cross-border transfer controls. Because this skill is specifically about combining audience signals across advertising ecosystems, the omission makes unsafe data-sharing practices more likely in real marketing operations.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.