ClawHub
Back to skill

Security audit

memo-quickstart-free

Security checks across malware telemetry and agentic risk

Overview

This local memory skill is mostly purpose-aligned, but it automatically persists user details and contains inconsistent network/API claims, so it should be reviewed before installation.

Install only if you are comfortable with the agent creating persistent local memory files from your conversations. Avoid storing passwords, financial details, secrets, or regulated data, and verify the npm package source before running the global install command.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

High
Confidence
95% confidence
Finding
The skill claims 'zero network requests' and 'no API key required' while also exposing a callback_url parameter, describing API integration, and documenting network/API failure modes. This contradiction can mislead users and agents into trusting the skill as purely local when it may transmit data externally, increasing the risk of unintended data disclosure.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill describes persistent writes to SESSION-STATE.json, MEMORY.md, and memories/ without a prominent upfront retention warning or explicit consent flow. In an agent context, silent persistence of user content can capture sensitive information unexpectedly and create privacy, compliance, and forensic exposure on the local system.

Ssd 3

Medium
Confidence
97% confidence
Finding
The workflow instructs the agent to persist user preferences, decisions, deadlines, and corrections by default. In practice, these records can contain sensitive personal, business, or security-relevant information, and default collection/storage without minimization or consent materially increases privacy risk and the chance of long-term leakage through local files, backups, logs, or later retrieval.

Ssd 3

Medium
Confidence
96% confidence
Finding
The examples repeatedly direct the agent to save user statements as memory records and confirm that saving occurred, normalizing persistent collection as a default behavior. In a memory skill with write and exec capabilities, this context makes the issue more dangerous because it encourages broad retention of conversational data that may later be exposed, misused, or searched beyond the user's expectations.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.