ClawHub
Back to skill

Security audit

Agent Memory Layer

Security checks across malware telemetry and agentic risk

Overview

This is a local agent-memory helper that intentionally stores long-term memories on disk, which is expected for its purpose but requires careful data handling.

Install only if you want agent memory to persist locally. Do not store secrets, regulated personal data, or sensitive business context unless you have reviewed the .agent_memory storage location and have a deletion or retention plan.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill is explicitly designed to persist short-term, long-term, and episodic memory, including user preferences and learned patterns, but provides no warning about privacy, retention, consent, or handling of user-related data. In an agent context, this can lead to silent accumulation of sensitive personal or behavioral data beyond user expectations.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
Automatic promotion of episodic memories into long-term storage increases persistence and broadens downstream reuse of data without any user-facing warning or safeguards. This is dangerous because transient interactions, failures, or personal details may become durable profile data, making privacy leakage and unexpected retention more likely.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The code persists long-term memory contents to disk automatically via `long_term.json` without any consent, disclosure, encryption, or access-control safeguards. In an agent-memory component, stored content may include sensitive prompts, personal data, credentials, or proprietary context, so silent persistence increases privacy and data-retention risk if the host is shared, compromised, or logs/backups are exposed.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.