Security audit

MLM Assistant

Security checks for vulnerabilities and agentic risk

Overview

The skill is transparent and local, but its MLM outreach and product-claim guidance needs review for privacy, consent, and misleading health or sanitation claims.

Review this carefully before installing. It appears local and not technically invasive, but users should avoid uploading unnecessary prospect data, should get consent before repeated outreach, and should not use the skill to make medical, disease, sanitation, earnings, or guaranteed-income claims without verified disclosures and applicable legal review.

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
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 97% confidence
Finding: The declared description presents a comprehensive MLM business assistant covering many conversational and operational functions such as prospecting, follow-ups, objection handling, team building, compliance, and duplication systems. The supplied code does not implement any of those capabilities. It only performs income projection calculations for Enagic or a generic MLM compensation model and prints the results. While income calculations are one listed aspect of the declared purpose, the code chunk’s actual primary purpose is much narrower than the declared assistant behavior. This is a material description-behavior mismatch because the implementation is a focused calculator utility rather than a general MLM business assistant.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation description is broad enough to trigger on common mentions of MLM companies or direct-sales topics, which can cause the skill to engage outside a clearly intended context. Overbroad activation increases the chance of unsolicited sales-script generation, manipulative outreach guidance, or handling sensitive business/personal scenarios without explicit user intent.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The skill encourages prospect categorization, outreach, follow-up sequencing, and conversation tracking without any privacy, consent, retention, or data-minimization guidance. In practice, this can lead users to process personal contact information and behavioral signals from social media in ways that are intrusive, non-compliant, or inconsistent with platform rules and privacy obligations.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This content presents product outputs for drinking, skin care, sanitizing, and disinfecting in a reference document without immediately pairing those uses with clear cautions that efficacy, safety, and regulatory status may be limited or unverified. In an MLM sales-assistant context, such statements can be reused by the agent as persuasive claims, increasing the risk of unsafe reliance, misleading health/sanitation representations, or compliance violations.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The selling points and demo guidance include body-related and quasi-scientific claims such as hydration, antioxidant properties, body pH balance, and test demonstrations that can imply medical benefit without an immediate caution about disputed science or limits of evidence. Because this skill is specifically designed to help prospecting and objection handling for MLM sales, the context makes the issue more dangerous: the assistant may systematically generate persuasive but misleading health-adjacent messaging at scale.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.