ClawHub
Back to skill

Security audit

mnemosyne

Security checks across malware telemetry and agentic risk

Overview

This is a local long-term memory skill, but it automatically stores conversation history and changes agent behavior without enough user control guidance.

Install only if you want an agent-wide persistent memory layer. Review the external installer before running it, confirm where memory files are stored, and make sure you know how to inspect, delete, disable, or back up stored conversation history and profile facts.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The manifest description says to use the skill when the user asks 'do you remember X' or complains the agent forgets, alongside broad conditions like needing recall of user facts, preferences, decisions, or conversation history. These triggers are common in everyday chat and do not define clear boundaries for when the skill should or should not activate, increasing the risk of unintended invocation.

Missing User Warnings

High
Confidence
97% confidence
Finding
The markdown states that the skill provides cross-session recall of user facts, preferences, decisions, and conversation history, and that memory survives restarts, but it does not warn users about the privacy and data-retention implications. Because this behavior affects user data and persists sensitive context locally, the skill description should explicitly disclose storage and retention behavior.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.