agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes examples and commands that pass credentials and secrets directly as CLI arguments or payloads (e.g., fill @e2 "password123", --proxy http://user:pass@..., set credentials user pass, --headers '{"X-Key":"v"}'), so an agent using it would be expected to embed secret values verbatim in generated commands/outputs.

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

• Third-party content exposure detected (high risk: 0.90). The skill's Navigation and Core workflow (e.g., the "agent-browser open " command and the "snapshot" / "get text" / "get html" actions) explicitly load and read arbitrary public web pages (http/https/file/etc.), so untrusted, user-generated content from those pages can be ingested and influence subsequent automated interactions.