ClawHub
Back to skill

Security audit

图表工具基础版

Security checks for vulnerabilities and agentic risk

Overview

This diagram skill mostly matches its purpose, but its broad automatic tool use and unclear privacy/API-key guidance need review before installation.

Install only if you are comfortable with an agent using shell execution for diagram work. Avoid giving it sensitive diagrams, prompts, or API keys until the publisher clarifies which operations stay local, which contact external APIs, and where secrets should be stored.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill claims that all FREE-version data remains local and is not uploaded to the cloud, but elsewhere states that some features require network access and external APIs. This can mislead users into providing sensitive diagrams, prompts, or metadata under a false privacy assumption, causing inadvertent data disclosure to third-party services.

Intent-Code Divergence

Low
Confidence
92% confidence
Finding
The documentation inconsistently says API keys may be stored in configuration files, then later says keys must not be hardcoded into code or configuration files. This ambiguity can lead users to place secrets in insecure locations, increasing the chance of accidental disclosure through source control, backups, or shared directories.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger condition is overly broad, covering general data analysis, reporting, statistics, and visualization requests rather than narrowly scoping diagram generation. In an agent context, this increases the chance the skill is auto-invoked for unrelated requests, leading to unnecessary command execution or external API usage beyond user expectations.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill description and quick-start flow do not clearly warn users that some functions may execute shell commands and contact external APIs. In a tool-enabled agent, missing this disclosure weakens informed consent and can cause users to unknowingly trigger risky operations in their local environment or send content off-host.

Static analysis

No suspicious patterns detected.