ClawHub
Back to skill

Security audit

Liuyao

Security checks across malware telemetry and agentic risk

Overview

This is a local Liuyao divination skill whose bundled references and helper scripts match its stated purpose, with only minor transparency notes.

Install only if you are comfortable with a divination skill that may run small local Python helpers, read its bundled reference files, and optionally append maintenance notes to its own log when manually invoked. Treat outputs about health, finance, investment, or major life decisions as entertainment or reflection, not professional advice.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill declares no permissions, yet its documented structure and behavior imply file reads from references/ and writes to logs/evolution_log.md via supporting scripts. Undeclared read/write capabilities weaken transparency and consent boundaries, making it easier for a seemingly harmless divination skill to access or persist data beyond user expectations.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The skill is presented as a divination assistant, but the referenced behavior includes inspecting skill files, counting content, generating improvement suggestions, and persisting audit-like logs. This mismatch is dangerous because users and reviewers may authorize the skill for spiritual/entertainment use while it performs maintenance-style analysis and file persistence unrelated to the declared purpose.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.