core-agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes plaintext credentials used as command-line arguments (e.g., agent-browser fill @e2 "password123"), which forces the agent to emit secret values verbatim in its outputs/commands.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses agent-browser open <url> / snapshot to fetch and parse arbitrary web pages (public web content supplied by the operating user), which can include outsider-authored free text that becomes LLM-readable in the snapshot output.