debugging

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill contains commands that explicitly read sensitive values (e.g., localStorage.getItem('token')) and capture HTTP requests/console logs which can include Authorization headers, cookies, or tokens, so the agent could receive and output secrets verbatim.

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflows show the agent navigating to and inspecting arbitrary web pages and resources (e.g., "browser-devtools-cli navigation go-to --url ...", "browser-devtools-cli content get-as-html", "o11y get-console-messages", "get-http-requests", and "run js-in-browser"), so it ingests untrusted public web/user-generated content that could contain instructions influencing subsequent actions.