ClawHub
Back to skill

Security audit

browser-testing-with-devtools

Security checks for vulnerabilities and agentic risk

Overview

The skill is not clearly malicious, but it asks for broad read/command authority while giving vague browser-testing instructions and unrelated API-credential guidance.

Review this skill carefully before installing. It may be usable for browser debugging, but only run it on sites and projects you explicitly authorize, avoid providing API keys unless the publisher documents a real need, and treat any command execution or elevated-permission suggestion as requiring separate confirmation.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger condition states the skill activates whenever the user needs 'related operations,' which is overly broad for a skill that exposes read and exec tools and can interact with browser-testing workflows. Ambiguous invocation criteria can cause the skill to run in inappropriate situations, increasing the risk of unintended command execution, unnecessary file access, or misuse against websites/environments that were not clearly authorized.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The summary and description contain a mandatory mixed-language presentation (Chinese and English) but do not state any user-selectable language or locale option. This can violate language/locale policy expectations because the skill imposes its communication format rather than offering opt-in or a documented regional justification.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The error-recovery section references API authentication failures and rate limits even though the skill is described as local/interactive browser testing via DevTools. Such contradictory guidance is risky because it can prompt users or agents to handle secrets, tokens, or remote service flows that are not actually required, obscuring the true trust boundaries of the skill and increasing the chance of unsafe credential handling.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill is presented as a Chrome DevTools/browser-testing capability, but these lines introduce API-key setup, API connection initialization, and generic API-calling steps unrelated to the stated function. This inconsistency can mislead an agent or operator into requesting, storing, or using credentials unnecessarily, expanding the attack surface and enabling accidental secret exposure or misuse of the skill in unintended contexts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.