ClawHub
Back to skill

Security audit

Auto Dealership Car Service Kit

Security checks for vulnerabilities and agentic risk

Overview

This is a marketing prompt kit with no executable payload, but it claims compliance while including review-gating and under-scoped targeted advertising guidance that users should review carefully.

Review this before installing if you plan to use it for live dealership marketing. Treat outputs as drafts requiring legal, privacy, platform-policy, and ad-disclosure review, especially for APR/price claims, SMS/email outreach, custom audiences, competitor-owner targeting, and review requests. Do not use the review-gating language as written.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The marketing copy makes specific price and APR claims but omits the required qualifying disclosures, stock-number references, and price-warning language that the same file later identifies as necessary under Nevada and federal advertising rules. In a dealership marketing skill, this is dangerous because users may copy these examples directly into customer-facing ads, creating compliance, deceptive-advertising, and enforcement risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The trade-in email implies the dealership used ownership or market-segmentation data about the recipient's vehicle without explaining the source, permission basis, or opt-out/privacy context beyond a generic CAN-SPAM footer. In a customer outreach skill, this can normalize privacy-invasive targeting and create legal and reputational exposure if operators rely on inferred or third-party vehicle-owner data without proper notice or consent.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · prompts/prompt3-reputation-reviews.md (reported line 53)May include surrounding context.

md
- Template A: Bad trade-in value complaint
- Template B: Bait-and-switch advertising allegation
- Template C: Mechanical issue after purchase
- Template D: Finance rate complaint ("they added products without asking")
- Template E: Service tech error (work done wrong)
- Template F: Warranty/gap dispute
- Template G: Racial/gender discrimination allegation (crisis-level; move to private immediately)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The prompt explicitly instructs sending the public Google review link only to customers who indicate they would rate the business 5 stars, while routing unhappy customers to a private channel. This is classic review gating, which can violate platform policies and consumer protection expectations by manipulating public reputation and suppressing negative feedback.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The retargeting instructions call for custom audiences based on website visitors and VDP viewers, which implies tracking, audience building, and behavioral advertising without any warning about consent, pixel/cookie disclosure, or platform-policy compliance. This is dangerous because it can encourage deployment of ad tech that processes browsing behavior and personal data without adequate notice or permission.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The prompt instructs creation of lead-nurture messages using personal lead data and inquiry details, but it does not require any privacy notice, consent validation, opt-out language, or lawful-basis check for CRM use. In a real dealership marketing workflow, this can normalize contacting individuals based on collected personal data without clear transparency or honoring consent, creating privacy and compliance risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The conquest campaign directs targeting of competitor owners and likely lease-expiration status, which relies on sensitive marketing segmentation and third-party or inferred ownership data, yet omits any requirement to verify lawful sourcing, consent, or permitted use. In context, this increases the risk of privacy violations, unlawful SMS/email outreach, and misuse of brokered audience data for highly targeted advertising.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.