ClawHub
Back to skill

Security audit

Memory Hygiene

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly coherent for memory cleanup, but it recommends an unguarded manual wipe and a recurring cron job that can repeatedly delete the user's vector memory.

Install only if you are comfortable manually controlling memory deletion. Before using the wipe or cron examples, back up `~/.clawdbot/memory/lancedb/`, verify the path, and avoid recurring deletion unless you have a tested reseed and recovery process.

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 (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill provides a direct destructive command (`rm -rf ~/.clawdbot/memory/lancedb/`) that permanently deletes all vector memory, but it does not explicitly warn about irreversible data loss, backup needs, or require confirmation. In a maintenance-focused skill, users are likely to copy-paste commands, so the omission materially increases the chance of accidental deletion of useful memories and service disruption until reseeding/restart occurs.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The cron job automates recurring destructive maintenance by scheduling a monthly wipe of the memory database, yet it omits safeguards such as warnings, backup steps, approval requirements, or validation of reseeding success. In context, automation makes this more dangerous than a one-time manual command because a misconfigured or no-longer-needed job can repeatedly erase memory without contemporaneous human review.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.