ClawHub
Back to skill

Security audit

English Learn Cards

Security checks across malware telemetry and agentic risk

Overview

This is a coherent English flashcard skill that stores study data locally and optionally looks up pronunciation audio online.

Install if you are comfortable running a local Python helper that writes a SQLite vocabulary database. Avoid --fill-audio or --audio-auto for sensitive words or phrases unless you are comfortable sending those headwords to Cambridge Dictionary for pronunciation lookup.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding
The skill documentation declares no permissions, but the described helper usage and referenced functionality imply access to environment variables and network resources. Undeclared capabilities are dangerous because reviewers and deployers may trust the skill to be local-only while it can reach external services or read runtime configuration, which weakens informed consent and security review.

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The stated purpose presents the skill as a portable SQLite/SRS flashcard tool, but the implementation reportedly also performs third-party scraping, external audio retrieval, and platform-specific behavior. This mismatch is risky because users may enable or deploy the skill without realizing it sends data to external sites, depends on unstable scraped content, or behaves differently across platforms, increasing privacy, reliability, and supply-chain exposure.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The command triggers are short, natural-language phrases like 'find <word>' and especially 'teach me', which can plausibly appear in ordinary conversation. In a chat-integrated agent, this can cause unintended skill activation, leading to unexpected database reads/writes or quiz state changes when the user did not intend to invoke the skill.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.