browser-use
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate documentation for the 'browser-use' automation library. It includes appropriate security guidance such as using
sensitive_dataplaceholders to prevent passwords from being sent to LLM providers in plain text, utilizingallowed_domainsto restrict the agent's movement, and advising users to git-ignore session files (auth.json). - [EXTERNAL_DOWNLOADS]: The skill mentions downloading Chromium via
uvx browser-use install. This is a standard dependency installation step for browser automation libraries and is considered safe within the context of the tool's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The documentation correctly identifies the risk of indirect prompt injection when the agent processes untrusted web pages and explicitly recommends setting
allowed_domainsand using sanitization practices as mitigations.
Audit Metadata