Back to skill

Security audit

Customer LTV

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent marketing-planning guide, but it should be reviewed because it encourages customer PII use and SMS/phone/ad outreach with inconsistent consent and data-minimization guardrails.

Install only if your team will apply its templates through your own privacy and compliance process. Before using it, restrict exports to the minimum customer fields needed, use approved storage, confirm privacy notice coverage, honor legal suppression lists, require channel-specific opt-in for all SMS/phone outreach, and review any Facebook/ad-audience uploads under your company policy.

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
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)

Context Leakage

High
Category
Data Exfiltration
Confidence
88% confidence
Finding

The instruction to 'Log conversation in CRM' encourages storage of potentially sensitive free-form call notes derived from personalized outreach, but provides no minimization, access-control, retention, or redaction guidance. In practice, staff may record more personal data than necessary, including inferred preferences or sensitive details, increasing exposure through internal over-sharing, retention creep, or downstream misuse.

Content

Scanner excerpt · references/output-template.md (reported line 302)May include surrounding context.

md
- Day 10 (Phone Call, if LTV >$[threshold]): Personal outreach from account manager or concierge
  - Script: "Hi [NAME], I'm [NAME] from [BRAND]. We noticed you haven't visited in a while. I wanted to personally check in—are you still [PRODUCT] or would something new be helpful?"
  - Offer: Email to follow with exclusive incentive based on conversation
  - Log conversation in CRM
- Day 12 (Email if No Phone Response): "Personal note from [TEAM]—let's find what works for you"
  - Softer tone; less salesy
  - Offer: Same 40% off; add concierge consultation (20-min call)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly instructs handling customer identifiers, email addresses, SMS messaging, and even phone-call outreach for segmentation and reactivation, but it provides no privacy, consent, or lawful-basis guardrails. In a marketing context, this omission can lead users to process personal data, target inactive customers, and escalate to intrusive outreach channels without checking opt-in status, regional rules, suppression requirements, or data-minimization obligations.

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 includes an email field in the minimum dataset structure for LTV analysis even though the analytical methods described do not require direct identifiers. Encouraging export of customer-level data with personal identifiers, without any minimization or handling guidance, increases the risk of unnecessary exposure, sharing, or retention of PII in spreadsheets and ad hoc workflows.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guide recommends exporting customer transaction data from Shopify or payment processors into CSV/Sheets workflows without any privacy or security warning. In the context of customer analytics, this is risky because such exports commonly contain PII and financial purchase history, and moving them into local files or spreadsheets broadens the attack surface and likelihood of accidental disclosure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The template recommends phone and SMS win-back outreach and even states 'Never suppress from any channel unless explicitly unsubscribed' for high-value customers, which encourages aggressive contact practices without a strong consent gate. If used as written, teams may contact individuals through regulated channels without verifying prior opt-in, quiet hours, do-not-call restrictions, or customer-impact safeguards, creating legal and reputational risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file provides fixed outreach copy in English for email, SMS, and phone scripts, but nowhere offers localization choice or states that the template is limited to an English-language audience. That creates a natural-language policy issue under the language/locale rule because the skill implicitly forces a single language without user opt-in or documented justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The template operationalizes customer segmentation, outreach channels, and behavioral data use without requiring an upfront privacy, consent, or lawful-basis check. In a marketing automation context, this can lead users to process personal data and launch outreach based on LTV/RFM profiling without confirming notice, consent, suppression rules, or jurisdiction-specific requirements.

Content

No source excerpt is available for this finding.