agent-browser
Fail
Audited by Snyk on Feb 15, 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 shows commands that embed secrets inline (e.g., fill "password123", set credentials user pass, set headers '{"X-Key":"v"}', cookies set name value), which requires the agent/LLM to include secret values verbatim in generated commands and outputs, creating exfiltration risk.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill clearly navigates arbitrary public web pages using commands like "agent-browser open " and ingests page content via "snapshot", "get text"/"get html", and "eval", meaning it will read untrusted third‑party (web/social/user‑generated) content that could contain indirect prompt injections.
Audit Metadata