ClawHub
Back to skill

Security audit

加密博弈论分析（免费版）

Security checks across malware telemetry and agentic risk

Overview

The skill mostly describes a crypto game-theory analysis workflow, but it requests shell execution and mentions callback URLs without clear limits or data-handling disclosure.

Review this skill before installing if you will analyze private protocol designs or business-sensitive material. Use it only where shell execution is disabled or explicitly approved, and avoid supplying callback URLs unless you understand exactly what data may be sent.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (4)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill advertises itself as an instruction-driven local analysis framework with no extra API key or external tooling, yet it accepts a callback_url for asynchronous completion. That creates an undocumented outbound data flow risk because user-supplied analysis content could be transmitted to an external endpoint, which is especially sensitive in an AI-agent context handling proprietary protocol details.

Intent-Code Divergence

Medium
Confidence
84% confidence
Finding
The documentation states the free version lacks external tool integration, but the presence of callback_url indicates an external integration path. This mismatch can mislead users into assuming analysis remains local when in fact outputs may be delivered off-platform, undermining informed consent and increasing the chance of unintended data disclosure.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
Documenting callback_url without warning about outbound transmission obscures that user content or generated analysis may be posted to a third-party endpoint. In a security-analysis skill, inputs can include sensitive design details, so silent network egress materially raises confidentiality and data-handling risk.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill declares exec capability but provides no warning, constraints, or safe-use guidance for command execution. In agent environments, exposing exec materially increases the chance that the skill could trigger local shell commands against user data or the host system, making this far more dangerous than a purely descriptive markdown skill.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.