Back to skill

Security audit

OpenClaw 浏览器自动化

Security checks for vulnerabilities and agentic risk

Overview

This is a browser automation instruction skill with powerful but disclosed web interaction features, and no bundled executable code or hidden persistence.

Install only if you want broad browser automation. Be careful with uploads, downloads, form submission, logged-in pages, and the optional Chrome profile because those actions can move local data or act through existing web sessions.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes very generic terms such as '浏览器', '浏览网页', and 'browser', which are likely to match many ordinary user requests and auto-invoke a powerful browser automation skill unnecessarily. Because this skill can navigate, click, upload, download, and execute page actions, accidental routing expands the chance of unintended external interaction or data transfer.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill advertises file upload/download and active web interaction capabilities without a prominent warning that these actions may transfer local data to remote sites or cause persistent external side effects. In practice, users or upstream agents may invoke these features without realizing that local files, authenticated sessions, or downloaded content are crossing trust boundaries.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
95% confidence
Finding

The trigger '截图' is extremely short and common, so it may activate this browser-control skill for unrelated screenshot requests. Since the skill exposes broader browser automation than simple image capture, an overly broad trigger can cause the system to select a more privileged tool than the user intended.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.