agentcore-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill requires the agent to emit tool calls containing a "text" field (e.g., browser_interact with action=type), so if asked to type a password/API key the LLM would need to include that secret verbatim in its generated command/output.

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

• Third-party content exposure detected (high risk: 0.95). Outsider free text is ingested when browser_navigate calls page.goto(url) and then page.evaluate(...) extracts clone.body.innerText/textContent from the runtime-fetched page at the user-supplied url (public/third-party web content).