ClawHub
Back to skill

Security audit

Canvas

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a coherent canvas helper, but it exposes remote WebView JavaScript execution and screenshots without enough safety scoping or privacy guidance.

Install only if you are comfortable letting agents control canvases on trusted OpenClaw nodes. Avoid using eval on untrusted pages, avoid navigating nodes to untrusted URLs, and treat snapshots as potentially sensitive because they may capture private content shown in the canvas.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly exposes an eval action that executes JavaScript in a remote node's WebView, but it provides no warning, trust boundary discussion, or restriction guidance. In this context, rendering arbitrary HTML/JS on connected Mac/iOS/Android nodes can enable code execution within the WebView context, data exfiltration from the displayed page, phishing-style UI spoofing, or abuse of any bridge-exposed capabilities.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documented snapshot action captures screenshots from connected nodes without any privacy or consent warning. Screenshots can expose sensitive application data, tokens, personal information, or internal dashboards visible on the node, especially because the skill is designed to operate across remote connected devices.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 88)May include surrounding context.

md
## Workflow

### 1. Create HTML content

Place files in the canvas root directory (default `~/clawd/canvas/`):

Static analysis

No suspicious patterns detected.