ClawHub
Back to skill

Security audit

Html To Markdown

Security checks across malware telemetry and agentic risk

Overview

This is a coherent HTML-to-Markdown helper whose URL fetching and file output are expected for its purpose, though users should run it only on intended URLs and paths.

Install only if you trust the external npm package. Before use, confirm the exact URL, input file, and output file. Avoid internal services, private metadata endpoints, authenticated pages, secrets, or paths you do not intend the agent to read or overwrite.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation examples are broad enough that an agent may invoke the skill for loosely related requests without clear user confirmation or scope boundaries. In a skill that can fetch remote URLs and write files, over-broad triggering increases the chance of unintended network access or filesystem modifications from ambiguous prompts.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The quick-start shows URL conversion as a normal operation but does not warn that this performs a live network fetch of remote content. In an agent setting, this can surprise users, expand the trust boundary, and enable unintended requests to external sites, including internal or sensitive endpoints if guardrails are weak.

Missing User Warnings

Low
Confidence
83% confidence
Finding
The file conversion example writes output to disk without warning the user that the command creates or overwrites local files. In an automated agent workflow, undocumented write behavior can lead to unintended file modification, clobbering of existing data, or confusion about where artifacts are stored.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.