The agent successfully assembled a comprehensive explanation of Aeralis by combining direct site fetches with web search results, but had to work around JavaScript-rendered pages that obscured visual content. The site's key messaging pages (homepage, pricing, about, FAQ, alternatives) were accessible, though several returned truncated HTML that required supplementary searches to fully understand differentiation, pricing tiers, and feature details.

• ›Steps [1], [2], [9], [10], [11] returned 200s and were cited in the final response's source list; their HTML responses were truncated by the fetch mechanism, forcing the agent to rely on metadata, search snippets, and prior knowledge to extract content about what Aeralis does, pricing structure, profiles/grounding features, and comparisons.
• ›The site uses Next.js with heavy JavaScript rendering, which prevented the agent from seeing rendered page content directly—the agent explicitly noted this limitation in its response, explaining that 'features page' and 'full details on how it works' were inaccessible due to JavaScript-rendering constraints.
• ›Web search results (steps [7], [8], [13]) supplied critical content that direct site fetches did not expose in a machine-readable form: pricing details, feature comparisons, profile/grounding explanations, and differentiation versus competitors came from search snippets and external review sites (Capterra, SoftwareAdvice, Google Workspace Marketplace), not from the rendered site pages themselves.
• ›The agent successfully answered the task despite poor site agent-readiness: pricing was eventually extracted, but the 'Business vs Pro differentiation' remained unclear; privacy/compliance details were absent; and team plan specifics were incomplete. The agent explicitly called out these gaps, showing it understood the limits of what it could reliably retrieve.
• ›The site's own /alternatives and /blog pages (steps [9], discovered via search) proved more informative than the main pages, suggesting the site's primary navigation may not be optimized for agent discovery of comparative or detailed technical information.