ClawHub
Back to skill

Security audit

Commentmd

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Markdown review helper whose disclosed browser, local server, and comments-file behavior fits its stated purpose, with no evidence of exfiltration or hidden persistence.

Install only if you are comfortable with a Markdown file being opened through a local browser-based review flow and with a comments JSON file being created beside the source document. Avoid using it on documents containing secrets or regulated data unless you have verified the referenced serve.py implementation and the local server behavior.

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 (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill metadata/description does not warn that invoking the skill will start a local HTTP server, open a browser, and write a comments JSON file adjacent to the reviewed Markdown by default. That omission can lead users or downstream agents to trigger network-facing behavior and local file writes without informed consent, which is especially risky in sensitive repositories or remote/headless environments.

Static analysis

No suspicious patterns detected.