ClawHub
Back to skill

Security audit

MindAR WebAR 图像识别

Security checks for vulnerabilities and agentic risk

Overview

The skill provides coherent WebAR development guidance, with a browser testing option that needs care.

Use the SwiftShader flags only for trusted local test content in a disposable browser profile without signed-in accounts, and keep the browser sandbox enabled. Bind the test server to localhost and stop it after testing. The skill supplies examples rather than a complete runnable toolchain.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

The browser launch instructions enable unsafe SwiftShader without disclosing its security implications.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The recommended --enable-unsafe-swiftshader flag explicitly opts into software WebGL execution with reduced security guarantees, potentially exposing the browser process to attacks through malicious graphics content or renderer vulnerabilities. Its use for local synthetic-camera testing is legitimate and reduces exposure, but the skill does not warn about the security tradeoff or restrict the browser to trusted test content; the flag alone does not establish an exploitable vulnerability.

Content

No source excerpt is available for this finding.