Security audit

AI-Powered Markdown to Word Converter (LaTeX, Mermaid & Auto Formatting)

Security checks across malware telemetry and agentic risk

Overview

This skill does what it says: it converts Markdown files by uploading them to md2word.com and downloading the Word result.

Install only if you are comfortable sending Markdown file contents to md2word.com. Avoid using it for secrets, private notes, customer data, regulated documents, or proprietary material unless that third-party service is approved for that data.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The function documentation claims a local format conversion, but the implementation sends the file to md2word.com for remote processing. This is dangerous because users may reasonably expect local-only handling and could unknowingly transmit sensitive markdown contents to a third party, creating confidentiality and compliance risks.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill explicitly states it will upload user-provided .md files to an external website, but it does not warn users that their content leaves the local/system boundary and is processed by a third party. This creates a real privacy and data-handling risk because users may submit sensitive documents under the assumption that conversion is handled locally or within the trusted platform.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The workflow describes automatic upload to and download from a third-party service without any privacy notice, consent checkpoint, or discussion of security implications. That makes the behavior more dangerous in context, because the transfer is automated and silent, increasing the chance that users unknowingly expose proprietary or sensitive markdown content to an external service.

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: The code uploads the user-supplied markdown file to an external website without any explicit warning, consent prompt, or trust boundary documentation. In an agent skill context, this is more dangerous because users may assume the skill operates locally while the file contents may contain secrets, proprietary data, or regulated information that is exfiltrated to a third party.

Known Vulnerable Dependency: requests==2.31.0 — 3 advisory(ies): CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi); CVE-2026-25645 (Requests has Insecure Temp File Reuse in its extract_zipped_paths() utility func)

Low

Category: Supply Chain
Confidence: 96% confidence
Finding: requests==2.31.0

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.