skills/browserbase/stagehand/browser/Gen Agent Trust Hub

browser

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses the vendor's 'browse' CLI for its intended purpose. It specifically instructs against installing new packages or using external network tools, which reduces the attack surface.\n- [PROMPT_INJECTION]: The skill processes untrusted external content from websites, which is a standard surface for indirect prompt injection in browser automation.\n
  • Ingestion points: Page content and accessibility trees via browse snapshot as described in SKILL.md.\n
  • Boundary markers: None provided in the instructions to delimit untrusted content.\n
  • Capability inventory: CLI execution of browse for actions like clicking and typing.\n
  • Sanitization: No sanitization is performed on the retrieved web content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 07:28 AM
Security Audit — agent-trust-hub — browser