safe-browser

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses the Playwright library and instructs the user to install the Chromium browser via npx playwright install chromium. This is a standard setup procedure for the browser automation tools used in the template.
  • [COMMAND_EXECUTION]: The skill provides instructions to run local shell commands (cp, npm install, node) to initialize and execute the demo application. These commands are standard for Node.js development workflows.
  • [DATA_EXPOSURE]: The provided template script includes functionality to load environment variables from local .env files, including a specific path (~/Developer/scratchpad/.env). This is used for managing API keys locally and does not involve exfiltrating data to external entities.
  • [INDIRECT_PROMPT_INJECTION]: As a web-browsing and scraping tool, the skill is naturally exposed to untrusted data from external websites. However, the skill author explicitly identifies page content as untrusted and implements a security boundary (CDP Fetch interception) to ensure all browser actions remain within the defined policy.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 05:54 AM
Security Audit — agent-trust-hub — safe-browser