ClawHub
Back to skill

Security audit

HTML设计工具

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly an HTML/CSS design helper, but it asks for broad command, file, and API/credential capabilities without enough scoping.

Review this before installing. It may be useful for HTML/CSS generation, but only grant command execution or provide API keys when you know exactly which command or service is being used. Do not assume the advertised cloud collaboration, permission controls, commercial licensing, or model fine-tuning features exist unless the publisher documents the actual service, storage, and access model.

SkillSpector

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 (5)

Description-Behavior Mismatch

Medium
Confidence
86% confidence
Finding
The skill is presented as an HTML/CSS design tool, but the documentation expands into generic API connectivity and operational guidance that is not clearly scoped to design generation. This kind of scope drift can cause an agent or user to grant broader capabilities than necessary, increasing the chance of unintended network access or misuse of credentials.

Description-Behavior Mismatch

Medium
Confidence
80% confidence
Finding
Claiming cloud storage, permission controls, and team collaboration introduces functionality not supported elsewhere in the manifest and can mislead users about where data is stored or shared. Misrepresentation of sharing features can lead to unsafe assumptions about confidentiality and access control.

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The documentation introduces command execution for a design-oriented skill without clearly limiting what commands may run or why execution is necessary. In a skill with an `exec` tool, vague execution guidance can enable arbitrary command use, local file access, or chaining with user-controlled input.

Context-Inappropriate Capability

Medium
Confidence
84% confidence
Finding
Generic instructions to initialize API connections and use credentials are broader than the stated purpose and do not identify specific endpoints, trust boundaries, or secret-handling controls. This can normalize unnecessary outbound access and increase the risk of credential exposure or abuse.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The skill description is broad and lacks clear activation boundaries, which increases the likelihood that an agent invokes it for tasks outside narrow HTML/CSS design work. Overbroad triggering is more dangerous here because the skill declares `read`, `write`, and `exec`, so accidental invocation can expose local resources or perform unnecessary actions.

Static analysis

No suspicious patterns detected.