ClawHub
Back to skill

Security audit

Chart Generator 2 0 0

Security checks for vulnerabilities and agentic risk

Overview

This chart-generation skill is not clearly malicious, but it asks for broad read/write/command authority and describes API, credential, file-processing, and command-execution behavior beyond a well-scoped SVG chart tool.

Review this skill before installing. It may be reasonable for simple chart generation, but only use it if you are comfortable granting command execution and file access to a loosely scoped markdown skill. Avoid giving it secrets or broad workspace access unless the publisher clarifies exact commands, files, API calls, and SVG output behavior.

SkillSpector

Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The skill is marketed as a chart/SVG generator, but later claims file processing, API integration, and command execution capabilities. This mismatch can mislead agents or users into granting broader trust and permissions than expected, increasing the risk of unsafe exec or data-handling behavior under the cover of a benign visualization tool.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The response schema describes a generic text-processing result rather than chart or SVG output, which contradicts the stated purpose. Ambiguous output contracts can cause downstream automation to mis-handle results, trust unexpected content, or permit hidden non-chart payloads where an SVG artifact was expected.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The documented inputs are generic text-processing fields like content, mode, and style instead of chart-specific parameters such as dataset, chart type, labels, and rendering options. This broadens interpretation of what the skill may do and creates room for prompt/command overreach or unsafe handling of arbitrary user content.

Vague Triggers

Medium
Confidence
76% confidence
Finding
Broad activation wording without precise trigger boundaries makes the skill eligible for invocation in loosely related contexts. In an agent environment, overbroad invocation criteria can route unrelated tasks into a skill that has read/write/exec tools, increasing the chance of unnecessary privileged actions.

Static analysis

No suspicious patterns detected.