ClawHub
Back to skill

Security audit

teaching-app-builder

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local teaching-page generator with disclosed HTML output and CDN use, with some ordinary caution needed around broad activation and unsafe Markdown examples.

Install this if you want an agent to create local interactive teaching HTML files. Expect it to write a .html file and for opened pages to load selected libraries from external CDNs; avoid rendering untrusted Markdown with the provided marked/innerHTML snippet unless it is sanitized.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger guidance is overly broad and explicitly says the skill should activate even when the user does not mention a teaching app, as long as the request sounds like making a webpage, demo, or visualization. That creates a real risk of unintended activation on generic requests, which can cause the agent to take file-writing actions or steer the workflow into HTML generation when the user may have wanted a simple answer or a different artifact.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The snippet parses Markdown and assigns the resulting HTML directly to innerHTML without any sanitization. If mdText can contain untrusted content, an attacker can inject raw HTML and potentially script-bearing payloads, leading to DOM XSS in the generated teaching app.

Static analysis

No suspicious patterns detected.