What problem does it solve? Manually interacting with websites for testing, scraping, or repetitive workflows is slow and error-prone. This Skill gives AI agents programmatic control of a real browser through a simple CLI, enabling navigation, form submission, authentication, and content capture without writing custom automation code. ## Core Features & Use Cases - Snapshot-Based Interaction: Get compact element references (@e1, @e2) from accessibility-tree snapshots, then click, fill, select, and scroll using those refs with minimal token usage. - Authentication & Sessions: Handle logins via auth vault, persistent profiles, session state files, or importing cookies from an existing Chrome session, including OAuth and 2FA flows. - Capture & Verification: Take annotated screenshots, save pages as PDF, record video sessions, and diff page states or screenshots to verify changes. - Use Case: Test a web app's signup flow by opening the page, snapshotting the form, filling fields with refs, submitting, and diffing the result against a baseline screenshot. ## Quick Start Ask the agent to open a website with agent-browser, take an interactive snapshot, fill in the login form, and capture a screenshot of the result.
