ClawHub
Back to skill

Security audit

Agent Memory Ops

Security checks across malware telemetry and agentic risk

Overview

This is a local memory-maintenance helper that reads project memory files and prints audit/digest suggestions, with no evidence of hidden execution, network access, or file mutation.

Install this only for workspaces where you want an agent to inspect local memory notes. Run it from the intended project root, review suggestions before changing `MEMORY.md`, and avoid using report/dedupe output in shared logs if memory files may contain sensitive content.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (1)

Vague Triggers

Medium
Confidence
80% confidence
Finding
The invocation guidance uses broad natural-language triggers such as '整理 MEMORY.md' and '做一次 agent memory audit' without explicit limits on what files may be scanned, what modifications are allowed, or when human confirmation is required. In an agentic system, vague triggers can cause the skill to activate in contexts the user did not intend, leading to overbroad reading of workspace files or unwanted memory promotion/editing, especially because the skill is designed to scan memory stores and curate durable facts.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.