chrome-devtools

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes examples that pass secret values verbatim (cookie JSON with "value":"abc123", a Bearer token in --token, and a test script typing a plaintext password), requiring the agent to handle and emit secrets directly, which is high-risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill autonomously fetches and processes arbitrary public web pages (e.g., navigate.js, evaluate.js which evals code in page context, aria-snapshot.js and snapshot.js which page.evaluate and extract DOM/ARIA refs, plus console.js/network.js) so untrusted, user-supplied websites can be read and their content used to drive interactions and decisions.