browser-skills

Signs into a site using credentials provided via `vars` (typically sourced from environment variables) or via a pre-saved Playwright persistent context. Never harvests or stores credentials. See docs/ethics.md.

Detects whether the current page contains a captcha challenge (reCAPTCHA, hCaptcha, Cloudflare Turnstile, GeeTest, etc.). Reports detection and stops. Never attempts to solve. See docs/ethics.md.

Extract structured data from an HTML table, optionally walking through paginated views. Returns a list of row dicts keyed by header cells. The demo skill (used to monitor conference paper lists).

Confirms that the current page has finished loading and is ready for interaction. Detects loading spinners, skeleton screens, and content placeholders. Run before extraction skills.

Fills a multi-step form (stepper wizard) by routing labeled fields to values from `vars`, clicking "Next" between steps, and submitting at the end. Pairs with the `searchable-dropdown` and `date-picker-widget` skills for compound widgets.

Uploads a local file to an HTML5 file input, or triggers a download and captures the resulting path. Composes with multi-step-form for file-attaching workflows.

Picks a date in a calendar-style date-picker widget — clicks the input, navigates months if needed, clicks the target day. The famous "vision agent failure case" that motivates this whole library.

Selects an option in a searchable / typeahead dropdown (combobox). Types to filter, then clicks the matching option. Common on city pickers, country selectors, package-search dropdowns.

Navigates to the next page in a paginated listing by clicking a "next" link/button or pressing the typical keyboard shortcut. Pairs with extract-table-pagination for multi-page data collection.

Types a query into a site's search input, submits, and optionally applies one or more filter facets (categories, price ranges, attributes). Common on e-commerce, doc-search, and listing sites.

Dismisses newsletter / "subscribe" / "sign up for emails" popups that appear after a short delay or scroll trigger on content sites.

Dismisses exit-intent popups ("Wait! Before you go!", "Don't leave without your discount", abandonment recovery overlays) that fire on mouse-leave or scroll-up patterns.

Scrolls a page to load additional content via infinite-scroll, with stop conditions (no-more-content, max iterations, or target selector visible). Used for feeds, lists, and search results that lazy-load.

Detects and dismisses blocking modal dialogs (generic confirmations, "are you sure", interstitials, signed-out prompts). Sibling to dismiss-newsletter-popup and dismiss-cookie-banner.

Detects and dismisses cookie consent banners on any site. Use on the first interaction after navigation to any page that may show a banner.