ClawHub
Back to skill

Security audit

记忆安全扫描

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent memory-security scanner, but it needs review because it combines sensitive file access, scheduled alerts, quarantine writes, and broad privilege guidance.

Install only if you are comfortable with the agent reading memory and workspace configuration files. Keep local mode as the default, review any remote LLM or callback configuration before use, do not enable scheduled monitoring unless you know where alerts go, and avoid running it with sudo unless a narrowly scoped administrative review justifies it.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The skill describes scheduled monitoring and alerting that can repeatedly read memory/workspace files and send notifications, but it does not present a prominent user warning about this ongoing access and outbound signaling. In an agent context, continuous scanning plus alert channels can expose sensitive workspace contents or metadata unexpectedly, especially if users assume the tool is only on-demand and local.

Ssd 4

Medium
Confidence
95% confidence
Finding
The FAQ advises using sudo or a '--sudo' option to overcome file permission issues during scanning, which normalizes privilege escalation for a tool that can read, write, execute, quarantine, and schedule tasks. In this skill context, elevated execution substantially increases the blast radius of mistakes, malicious memory content interactions, or implementation flaws, potentially leading to unauthorized file access or destructive system changes.

Static analysis

No suspicious patterns detected.