ClawHub
Back to skill

Security audit

Hekouwang Claude Md Doctor Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed CLAUDE.md auditing and repair helper with read-first behavior and confirmation before edits.

Install if you want a CLAUDE.md auditor that can also help rewrite project instruction files. Before accepting fixes, review proposed edits carefully, especially changes to .claude/settings.json hooks or MEMORY.md because those can influence future agent sessions.

SkillSpector

By NVIDIA
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 (3)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The invocation phrase is broad enough that ordinary requests like 'check my CLAUDE.md' or similar audit wording could trigger the skill unintentionally. In an agent environment, this can cause unexpected document analysis or downstream actions, especially because the skill advertises that it may later apply fixes, increasing the chance of unintended workflow execution.

Missing User Warnings

Low
Confidence
82% confidence
Finding
The README states that the tool can 'apply the fixes for you' near the primary usage instructions, but does not place an equally prominent warning that this may modify project files. That can mislead users into invoking the skill under the assumption it is purely diagnostic, creating a risk of unintended file changes in repositories or CI-like workflows.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger language is extremely broad, covering nearly any request to evaluate, review, or optimize a project's CLAUDE.md. In an agent environment, overly permissive activation boundaries can cause the skill to auto-trigger in unintended contexts, leading to unnecessary file inspection, surprising behavior, or interference with other skills and user intent.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.