Complex web application scenarios for testing AI agent reasoning and navigation capabilities.
E-commerce admin dashboard with product management, orders, customers, and settings. Tests navigation, form filling, filtering, and multi-page workflows.
Comprehensive test page covering all CDP tool operations: clicks, inputs, dropdowns, checkboxes, iframes, modals, tabs, and dynamic content.
HR employee portal with employee management, time-off requests, payroll, and reports. Tests complex form workflows and data extraction.
Banking portal for testing error categorization. Agent must navigate failure scenarios and return the correct error_category for each type of error.
Resident portal for rent payments with payment methods, autopay settings, payment history, and lease management.
Property management portal with unit directory, inspection history in paginated modals, and caseworker directory. Tests CSV extraction from popups and multi-page tables.
Test agent orchestration patterns with parallel sub-agent delegation.
Large insurance application form with 4 independent sections. Ideal for testing parallel sub-agent delegation where each section can be handled by a separate agent.
Process 10 employee onboarding tasks. Each employee requires the same 3-4 click workflow. Tests ability to spawn sub-agents for each iteration of a repetitive loop.
Realistic site replicas for testing PAD switching, form filling, and end-to-end scraping workflows.
Generic PAD switching flow: login, select account, fill banking details (institution, transit, account number), submit, and capture confirmation screenshot.
Quebec-style utility company replica with Pre-authorized Debit switch flow. Supports single account, multi-account list, and portfolio selection flows.