The agent successfully assembled a comprehensive explanation of KeeperHub by combining fetches from the main marketing site (keeperhub.com) with web search results, since the docs subdomain (docs.keeperhub.com) returned incomplete or 404 responses. The marketing pages and comparison resources provided clear positioning, pricing tiers, and differentiation; however, the agent had to work around inaccessible or fragmented technical documentation to fill gaps about workflow mechanics, wallet custody, and SLA details.

• ›The agent's most valuable fetches were the main site pages (homepage, pricing, integrations, compare) rather than docs.keeperhub.com—docs pages returned 404s or truncated HTML, forcing reliance on marketing collateral and web search snippets to construct the answer.
• ›Pricing and positioning were clearly published on keeperhub.com/pricing and keeperhub.com/compare (step [7], [11]) as structured pages; the agent cited these directly. However, operational details (gas credit mechanics, SLA specifics, custodial vs. non-custodial wallet handling) were not present in fetched content and had to be flagged as unclear or missing.
• ›The agent discovered most resources via prior knowledge guesses and web search (steps [5], [10], [12], [15] yielded URLs) rather than by navigating site structure from docs.keeperhub.com. The docs domain appears to be a documentation portal with broken/incomplete pages, while the primary marketing domain (keeperhub.com) served as the authoritative source.
• ›Agent-readiness is mixed: high-level positioning, use cases, and competitive comparisons are published as human-readable web pages suitable for scraping; however, technical integration details, workflow semantics, and pricing execution mechanics lack depth in accessible form and would require direct API documentation or human contact.