ClawHub
Back to skill

Security audit

Agent Memory Hierarchy

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only memory organization skill, with disclosed local memory-file edits and no code execution or network behavior.

Install if you want an agent to reorganize its long-term memory. Before running a migration, make a backup, review the proposed facts.yaml and MEMORY.md changes, and avoid storing actual secrets or sensitive personal details unless you intentionally want them retained.

SkillSpector

By NVIDIA
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
Confidence
88% confidence
Finding
The skill instructs the agent to rewrite and relocate memory content across multiple files, but it does not require confirmation, backup, diff review, or data-preservation checks before modifying user memory. In practice, this can cause unintended overwrites, lossy migrations, or corruption of long-term agent state, especially because the instructions encourage removing material from the original MEMORY.md after migration.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.