chrome-devtools

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes examples that pass sensitive values (e.g., passwords) directly on the command line (e.g., node fill.js --selector "#password" --value "secret"), which requires the agent to include secrets verbatim in generated commands/outputs and thus poses an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can be ingested when the required runtime workflow navigates to a user-supplied URL and then extracts page-authored text (e.g., scripts/snapshot.js uses page.evaluate() to read el.textContent from the loaded page, and scripts/console.js captures msg.text() from the page’s console), which is then emitted as JSON to the agent/LLM context.