ClawHub
Back to skill

Security audit

💾 Complex Memory Manager

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local memory-management skill, with privacy caveats but no evidence of hidden code, network access, or deceptive behavior.

Install only if you want agents to maintain cross-session local memory. Treat the internal tier as weak obfuscation, not secure encryption, and do not store secrets, PII, account data, or sensitive personal details in it. Periodically inspect or clear the memory directories if retention matters.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are extremely generic for a memory-writing skill, so ordinary user language like 'remember this' or 'save this for later' can invoke persistent storage behavior without clear, deliberate consent. In the context of a cross-session memory manager, this increases the chance of over-collection, accidental retention, and unintended writes of sensitive or personal information.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill performs file-system-affecting actions such as writing memory, moving aged entries to archives, deleting entries containing PII, and compacting or merging files, but the description does not prominently warn users that these operations will occur. For a shared component used by other skills, this lack of disclosure can lead to surprising persistence, deletion, or relocation of data and weakens informed consent and auditability.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.