agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt includes examples that pass plaintext secrets directly in CLI commands (e.g., agent-browser fill @e2 "password123" and chained commands embedding credentials), which would require the LLM to include secret values verbatim in generated output and thus creates exfiltration risk.

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md workflow and command reference explicitly direct the agent to open arbitrary URLs (agent-browser open ) and then snapshot/get text from pages (agent-browser snapshot -i, agent-browser get text), meaning the agent fetches and interprets untrusted third‑party web content which can directly drive subsequent actions.