browser-testing-with-screenshots

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's examples and workflow instruct embedding plaintext credentials directly into commands (e.g., document.activeElement.value = "password123") and reference tools that expose cookies/session data, which requires the agent to handle or output secrets verbatim and thus creates an exfiltration risk.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill requires cloning and installing code from https://github.com/badlogic/agent-tools and then runs the fetched browser-*.js scripts (e.g., browser-start.js, browser-eval.js) at runtime, so external code is fetched and executed and is required for the skill to operate.