agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The examples and commands show entering and saving credentials directly (e.g., agent-browser fill @e2 "password123" and saving auth.json), which requires the LLM to include secret values verbatim in generated commands/outputs and thus risks secret exposure.

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

• Third-party content exposure detected (high risk: 0.95). The required runtime workflow uses agent-browser open <url> and agent-browser snapshot to fetch and parse arbitrary web page content (outsider-authored public web text) into the agent’s LLM context via the accessibility tree / interactive elements.