tanner-website-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs the agent to read and report network/response headers and notes that a real Chrome session may include logged-in tabs and cookies, which can contain session tokens or other sensitive header values that the LLM would output verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required workflow explicitly loads and parses arbitrary outsider-authored live web pages from user-supplied URLs (raw HTML/text/meta/JSON-LD and bot-UA responses) into the agent’s context, creating an indirect prompt-injection exposure path.