ClawHub
Back to skill

Security audit

Excalidraw Architect

Security checks for vulnerabilities and agentic risk

Overview

This Excalidraw diagramming skill is purpose-aligned, with the main caution that it can change the current Excalidraw canvas when used.

Use this skill when you want an agent to draw or revise architecture diagrams in Excalidraw. Because it reuses an existing board and can rewrite scene elements, run it on a new or backed-up canvas if the current board contains important work.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The declared description suggests a skill that works directly in Excalidraw from natural-language requests and can draw, update, clean up, or restructure diagrams. The supplied code is instead a standalone generator that converts a specific JSON schema into Excalidraw element JSON and writes it to disk. It creates titles, rectangles, text labels, and arrows from explicit coordinates and node/edge definitions. There is no browser automation, no API calls to excalidraw.com, no natural-language parsing, and no logic for modifying an existing diagram. While the output is related to Excalidraw architecture diagrams, the implemented behavior is materially narrower and different from the declared purpose.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill instructs use of browser evaluation and references local scripts/resources, but it declares no explicit tool scope or permission boundaries. That creates an authorization gap where an agent may use broader file read/write capabilities than intended, increasing the chance of unnecessary local file access or misuse if the skill is invoked in a permissive environment.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file includes a JavaScript snippet that changes the current Excalidraw scene via api.updateScene and adjusts viewport behavior with scrollToContent, but the surrounding documentation provides no warning that it will overwrite or alter the user's current canvas state. For markdown files, user-affecting behavior that can impact user data or system state should be disclosed explicitly.

Content

No source excerpt is available for this finding.