agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's examples and commands show plaintext credentials (e.g., "password123") and use CLI actions like agent-browser fill and cookies set that embed secret values verbatim, so an agent would need to output secrets directly.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md describes a runtime path where the agent uses agent-browser open <url> and then agent-browser snapshot -i / agent-browser get text/html to ingest page accessibility tree and element text/HTML from whatever URL is opened (public/outsider-authored web content), which can include free-text prompt-injection payloads.