ClawHub
Back to skill

Security audit

L6 Learning Accelerator

Security checks across malware telemetry and agentic risk

Overview

This is a local learning helper with no hidden network, credential, shell, or autonomous behavior found, though users should be careful with its broad triggers and optional file import/export helpers.

Install if you want local time-aware retrieval and learning progress tracking. Use explicit prompts when invoking it, keep import/export paths inside a workspace directory, avoid importing untrusted JSON, and remember that exported progress files may contain learning notes or session details.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The module documentation claims storage is in-memory, but the implementation also exposes import and export functions that read from and write to arbitrary filesystem paths supplied at runtime. This mismatch can cause callers or reviewers to underestimate the file system access the skill has, increasing the chance that untrusted input is passed into these functions and leading to unintended file overwrite, data exposure, or unsafe file reads.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list contains broad phrases like 'learning accelerator', 'accelerated learning', and 'learning boost' that are likely to appear in normal user conversation, which can cause unintended activation of the skill. In a skill that performs memory retrieval and progress tracking, accidental invocation increases the chance of exposing stored context or recording user activity without clear intent.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The description advertises progress tracking and memory-based retrieval but does not warn users that their learning data may be stored or persisted. Because the skill references a configurable 'memory_path' and progress reports over time, users may unknowingly provide data that is retained, creating privacy and consent risks.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.