What problem does it solve? AI agents and automation scripts need a token-efficient way to interact with websites without parsing raw HTML or brittle CSS selectors. This Skill provides a CLI that converts pages into compact snapshots with element refs, enabling reliable clicking, filling, and extraction. ## Core Features & Use Cases - Snapshot-Based Interaction: Get interactive element refs (@e1, @e2) and use them to click, fill, select, and scroll without writing selectors. - Session & State Management: Run parallel isolated sessions, save authentication state, and reuse logins across runs. - Capture & Extraction: Take screenshots, save pages as PDF, record video, and extract text or structured data from any page. - Use Case: Log into a web app once, save the auth state, then run a daily script that loads the state, navigates to a dashboard, and extracts the latest metrics as text. ## Quick Start Use the agent-browser skill to open https://example.com/login, fill in the email and password fields, submit the form, and take a screenshot of the resulting dashboard.
