The agent assembled a partial but useful explanation of MURSAIN by combining fragments from two successfully fetched pages (dehumidifier guide and capillary rise article) with inferred domain knowledge and web search context. Critical information — pricing, detailed technical process, and competitive differentiation — was either absent from the site or inaccessible due to heavy client-side JavaScript rendering, forcing the agent to acknowledge gaps rather than deliver a complete answer.

• ›Steps [12] and [18] returned HTML with client-side JS placeholders but no rendered content text. The agent extracted what it could about MURSAIN's service area and problem types (rising damp, condensation, mold), but could not access detailed explanations of their process or methodology from the fetched pages.
• ›The agent discovered the two successful pages ([12], [18]) through web search routing (steps [10] and [16]), not through site navigation. Direct attempts to reach /en/pricing, /en/services, and /en/about returned 404s, signaling that the site either has minimal structure or routes everything through the homepage (client-side rendered).
• ›The site is poorly agent-ready: no pricing page, no dedicated services/process explanation, no structured metadata, and heavy reliance on client-side rendering that stripped all human-readable content from the fetches. The agent had to rely on prior knowledge (29%) and web search (29%) to fill gaps that should have been on-site, and even then could only provide partial answers on pricing and technical differentiation.