agent-browser
Fail
Audited by Snyk on Mar 8, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The prompt includes examples that put secrets directly into CLI commands (e.g., --headers '{"Authorization": "Bearer token123"}' and fill commands with "password123"), which requires the agent/LLM to emit secret values verbatim in its output.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to open arbitrary external URLs (e.g., "agent-browser open ") and to snapshot/get text/find elements from pages ("agent-browser snapshot", "get text", "find text"), meaning the agent fetches and interprets untrusted public web content that can directly influence subsequent clicks, fills, and navigation.
Audit Metadata