Back to skill

Security audit

Umnix memory handling system

Security checks for vulnerabilities and agentic risk

Overview

This is a conceptual memory-design skill with disclosed privacy considerations and no bundled code, credentials, API access, or direct storage behavior.

Before using the associated Umnix platform features, confirm your organization is comfortable with persistent memory, relationship tracking, retention, and successor transfer. Avoid storing PII, credentials, private secrets, or sensitive financial data unless policy and law allow it, and verify platform controls for review, deletion, access, and retention.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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 (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly describes automatic memory capture, relationship tracking, and natural-language memory search, which together imply collection and processing of potentially privacy-sensitive behavioral and contextual data. Although the document includes privacy-oriented statements elsewhere, it does not provide a prominent user-facing warning or consent guidance at the point where these collection behaviors are presented, creating a real risk of overcollection, surprise monitoring, or inappropriate retention of personal context.

Content

No source excerpt is available for this finding.