The agent successfully gathered enough information to explain what Prana Tree does, who it serves, and how it differentiates itself, but encountered a critical gap: the site does not publish pricing information. The agent assembled its answer from the homepage, services pages, and an about page, all of which loaded successfully and contained descriptive content about service offerings and credentials. However, the site's reliance on React dynamic content and the absence of a dedicated pricing page meant the agent could only report that pricing is unavailable online.

• ›Step [1] (homepage) and step [3] (services) returned full service descriptions, certification claims ($10m public liability, NZQA Level 4), and service area boundaries—core identity content was present and accessible.
• ›Step [10] (/services/tree-care) provided additional service detail (pruning types, health assessments, emergency response) that the agent used to build comprehensive service descriptions.
• ›Step [4] (pricing page) returned a 404, and the agent correctly identified this as a deliberate design choice (free-quote model) rather than a bug—the site surfaces pricing philosophy through service pages, not a dedicated pricing resource.
• ›The site does not differentiate itself vs. competitors (no comparative positioning statements found); the agent relied on prior knowledge and search results to infer that this is typical for local service providers.
• ›React dynamic loading truncated fetched page content, limiting the agent's ability to extract gallery details or full team information—the site's rendering strategy made it partially opaque to fetch-based inspection.
• ›The agent had to use web search ([8]) to validate service scope and locate comparable competitors, indicating the site does not explicitly position itself against alternatives.