ClawHub
Back to skill

Security audit

Claude Memory Graph Audit

Security checks for vulnerabilities and agentic risk

Overview

This skill is a scoped Claude memory health-audit tool with a read-only audit script and documented manual repair guidance for broken links.

Install if you want Claude to inspect and help maintain your local Claude memory graph. Run the audit script freely for read-only checks, but before applying repair commands or adding links, preview the affected files and keep a backup or diff because those changes alter persistent memory content.

SkillSpector

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
90% confidence
Finding
The skill includes direct in-place edit commands (`sed -i`) that modify user memory files immediately, but it does not require explicit confirmation, backup creation, or a dry-run before changing user data. Because the skill is designed to operate on a user's persistent knowledge store, an incorrect replacement or broad match could silently corrupt links across many files and degrade future recall behavior.

Static analysis

No suspicious patterns detected.