ClawHub
Back to skill

Security audit

The Spaced Repetition Systems for Agents

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed local spaced-repetition review tool, but users should supervise what it stores and any memory changes it suggests.

Install only if you want a local SRSA-style memory review workflow. Avoid putting secrets or sensitive personal data into cards, review proposed memory add/delete/update actions before applying them, and use card override or removal only when you intentionally want to change stored review material.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The declared description frames the skill as a review/grading aid, but the documented commands also create, modify, and delete cards and persist review data in local storage. This mismatch can cause users or orchestrators to authorize the skill under a narrower trust model than its actual behavior, leading to unintended data modification and retention.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to add, delete, or update memory about user preferences and corrections without requiring consent, scoping, or retention safeguards. Because this concerns user-related memory, the skill can normalize silent persistence of personal or behavioral data and accidental corruption of stored user context.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.