agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's examples and commands require embedding passwords/credentials directly in CLI arguments (e.g., agent-browser fill @e2 "password123", agent-browser set credentials user pass), which forces the LLM to output secret values verbatim and risks secret exfiltration.

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can be ingested when agent-browser open <url> navigates to a public/attacker-controlled page and agent-browser snapshot/get text returns page-derived readable text (e.g., DOM text) into the agent’s LLM context.