ClawHub
Back to skill

Security audit

Memory Association

Security checks across malware telemetry and agentic risk

Overview

This skill is a local memory-recall helper that deliberately reads and updates OpenClaw memory files, with no evidence of hidden network, credential, or destructive behavior.

Install this only if you want OpenClaw to consult and update local memory during normal tasks. Before installing, review your OpenClaw memory and .learnings files for secrets or sensitive client data, and treat this as read/write memory automation rather than a read-only search helper.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
81% confidence
Finding
The skill instructs the agent to read from `references/MEMORY_INDEX.md` and write to memory files and learning files, but it does not declare corresponding permissions. Undeclared read/write capability is dangerous because it expands effective access beyond what a reviewer or policy system can reliably see, increasing the risk of silent data exposure or unauthorized persistence.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger conditions are extremely broad: the skill should run before any new task, during analysis, while solving problems, on memory-related questions, and during session startup. Over-broad automatic invocation can cause unnecessary access to prior memory on unrelated tasks, increasing the chance of privacy leakage, irrelevant context injection, and unnecessary file operations without clear user intent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.