ClawHub
Back to skill

Security audit

Excalidraw Diagram Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a diagram generator whose file creation and optional icon-library instructions are mostly aligned with that purpose, with some scope and documentation caveats.

Before installing, expect this skill to create .excalidraw files and, if you use icon libraries, to ask you to download library files and run local setup/helper commands. Review any generated command paths first, especially because the artifact references helper scripts that were not included in the inspected package.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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 (4)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The manifest description says to use the skill when asked to "create a diagram," "make a flowchart," or "visualize a process," which are relatively broad activation phrases. The file does not provide exclusion conditions or negative examples to clarify when such requests should not invoke this skill, increasing the chance of unintended activation.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The manifest advertises a narrow content-generation skill, but the body documents additional capabilities for library setup, file placement, script execution, and post-generation diagram editing. This mismatch can cause an agent or operator to grant the skill broader authority than intended, increasing the chance of unsafe tool use and making security review harder because the actual behavior exceeds the declared scope.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill’s stated purpose is to generate Excalidraw JSON from natural language, but these instructions expand its behavior into running shell and Python commands that create directories and modify local files. That broadens the trust boundary from text generation to filesystem manipulation, creating unnecessary risk of unintended file changes, misuse of local tooling, or abuse if a user request is crafted to steer script execution or path selection.

Vague Triggers

Low
Confidence
84% confidence
Finding
The "When to Use This Skill" section lists several example requests, but it does not define non-matching cases or limits on trigger scope. Because phrases like "show the relationship between" and "diagram the workflow of" can arise in ordinary explanation requests, the activation criteria remain somewhat ambiguous.

Static analysis

No suspicious patterns detected.