ClawHub
Back to skill

Security audit

Taoism

Security checks across malware telemetry and agentic risk

Overview

This is a static Taoism reference skill with disclosed extra profile and promotional content, but it does not run code, access data, or change anything on the system.

Install only if you want a Taoism/Daoism reference skill and are comfortable with it also containing Chen Lang profile details and OpenClaw promotional material. A cleaner version would split the profile and project content into a separate clearly named skill, but the current artifact does not show unsafe system access or malicious behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill is presented as a Daoist culture information skill, but it embeds unrelated personal profile and project promotion content. This creates a scope mismatch that can cause the agent to disclose or promote personal/business information when users only asked for cultural content, increasing privacy, trust, and prompt-routing risk.

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
Publishing personal contact details and unrelated promotional material inside a cultural reference skill is unjustified and can lead to unsolicited disclosure, spam, profiling, or covert promotion through the assistant. In skill ecosystems, such embedded data may be surfaced to users unexpectedly, making the leakage more dangerous than a normal static document.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation description is broad enough to trigger on many common Taoism-related mentions, which increases the chance that the assistant loads this skill in loosely related conversations. Because the skill also contains unrelated personal and promotional content, broad triggering amplifies the risk of irrelevant disclosure and covert advertising.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The markdown exposes personal contact and profile information without warning or consent context, which can result in privacy leakage and unwanted contact. In an agent skill, this is especially risky because the assistant may repeat or prioritize this data conversationally, extending its reach beyond passive publication.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.