ClawHub
Back to skill

Security audit

Memory Audit — DDIA+DDD System Health Check

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent memory-audit tool, but it can direct broad edits to agent memory and behavior files without clear user approval boundaries.

Use this skill only when you intentionally want a memory-system maintenance pass. Run the baseline and diagnosis as read-only first, then require a proposed diff and explicit approval before any repair, AGENTS.md/MEMORY.md update, cron setup, or trace-logging integration.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill clearly instructs reading and modifying workspace memory artifacts and running local scripts, yet it declares no permissions. This creates a transparency and policy-enforcement gap: an agent or reviewer may underestimate the skill's access needs, and the skill can perform file read/write operations without explicit consent boundaries.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The trigger conditions are broad and subjective, such as memory 'feels stale' or periodic health checks, which can cause the skill to activate during ordinary discussion rather than an explicit maintenance request. In context, this is risky because the skill's workflow includes auditing, executing fixes, and updating project files, so over-broad activation can lead to unintended modification of workspace state.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill directs the agent to execute scripts, apply fixes sequentially, update multiple memory files, and record traces, but it does not warn users that these actions may alter workspace data. This is dangerous because an audit framed as maintenance can silently become a write-capable remediation workflow that changes project records, potentially causing data loss, drift, or unauthorized edits.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.