ClawHub
Back to skill

Security audit

Smart Memory Manager

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed memory-management skill whose main risk is local retention of user-provided memories, not hidden malicious behavior.

Install only if you want the agent to keep searchable memory. Avoid storing secrets, regulated data, or sensitive personal details, and use a dedicated memory file path you can review and delete. Treat any save/load or persist=true action as durable local retention.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger keywords are extremely broad terms such as "memory," "smart," "manager," and "agents," which are likely to match ordinary user conversation and cause unintended auto-activation. In an agent environment where the skill exposes read/exec capabilities and supports persistence, accidental invocation can lead to unwanted file operations, memory retention, or context handling beyond what the user intended.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documentation advertises persistent storage and restart survival without clearly warning users that conversation-derived memory may be written to disk and retained outside the live session. This creates a realistic risk of users or downstream agents storing sensitive personal, proprietary, or regulated data without informed consent or retention controls.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The save/load examples normalize writing memory to arbitrary local files but omit warnings about overwriting existing files, long-term retention, path safety, and handling of sensitive content. Because the skill is positioned for autonomous agent workflows, examples strongly influence implementation and may lead operators to persist user data insecurely by default.

Ssd 3

Medium
Confidence
88% confidence
Finding
The skill explicitly promotes storing user-specific conversation details and summaries as memory, which increases the chance of collecting and retaining personal or sensitive behavioral information beyond the immediate interaction. In the context of an agent memory manager with disk persistence and semantic retrieval, this can amplify privacy risk by making sensitive facts easier to retain, search, and reuse later.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.