agent-browser
Fail
Audited by Snyk on Feb 24, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 0.90). The skill includes examples and command patterns that embed literal passwords and form values (e.g., agent-browser fill "SecurePass123!" and subprocess.run([... row["password"] ...])) so an agent following it may be required to output secret values verbatim.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill explicitly navigates to and snapshots arbitrary public URLs (e.g., "agent-browser open " and "agent-browser snapshot -i" in SKILL.md and the scripts/browser_test.py which accepts --url and opens external sites), parses page text/JSON, and uses that untrusted page content to drive decisions and follow-up actions (tests/checks/clicks), so untrusted third-party content can influence agent behavior.
Audit Metadata