ClawHub
Back to skill

Security audit

Lithtrix Memory Identity

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent memory integration, but it encourages automatic long-term storage of user context in an external service without clear consent, sensitivity, retention, or deletion controls.

Install only if you are comfortable with the agent sending selected conversation context and preferences to Lithtrix for persistent storage. Before use, set clear rules to store only non-sensitive information, avoid secrets or regulated data, and ask Lithtrix or the publisher how memories can be reviewed, deleted, and retained.

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
96% confidence
Finding
The skill is explicitly designed to persist user context across sessions, but it provides no privacy notice, retention policy, or guidance to avoid storing sensitive personal data. In a memory skill, this omission is dangerous because agents are encouraged to collect and retain cross-session user information by default, creating risk of unauthorized profiling, over-collection, and privacy violations.

Missing User Warnings

High
Confidence
99% confidence
Finding
The prompt instructions tell the agent to automatically store summaries of meaningful tasks and user preferences after interaction, without requiring consent, sensitivity checks, or user review. This creates a direct pathway for exfiltrating personal, confidential, or regulated information into an external third-party memory service as part of normal operation.

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The example normalizes carrying forward a user's language preference into future sessions automatically, even though that preference is personal data and the user was not shown consenting to persistent storage for later use. While less severe than storing arbitrary task summaries, it still encourages silent cross-session profiling and behavior personalization without transparency.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal