ClawHub
Back to skill

Security audit

Mnemo Cortex

Security checks across malware telemetry and agentic risk

Overview

This is a coherent setup guide for local persistent agent memory, with privacy and sharing risks users should understand before enabling it.

Install only if you want agent memories to persist beyond one session. Review the referenced repository before running install commands, avoid saving secrets or regulated data, confirm which agent IDs can share memory, and learn how to inspect or delete stored memories before using it on sensitive projects.

SkillSpector

By NVIDIA
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 (1)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly promotes persistent cross-session recall and multi-agent shared memory, but it does not present a clear privacy warning or consent checkpoint before enabling these capabilities. This can lead users to store sensitive prompts, decisions, or project data in a durable local service that may later be queried by other sessions or agents, increasing the risk of unintended retention and disclosure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.