What does oaksun.studio do and who is it for? Explain it back to me.
- previous resource20%
- web search30%
- prior knowledge50%
The agent successfully pieced together what OAKSUN Studio does, who it serves, and how it differs from alternatives, but worked around significant gaps in the site's transparency—most critically, no pricing information was publicly available. The site's HTML was heavily compressed on direct fetches, forcing the agent to rely on web search results and external profiles (Awwwards, Contra) to construct a complete picture. While the agent delivered a coherent, well-reasoned answer grounded in actual content, the original site itself did not expose key business details (pricing, target audience clarity, package options) in machine-readable or easily fetchable form.
- ›Step [1] and [6] (homepage and services page) returned heavily truncated HTML that prevented direct extraction of detailed service descriptions, pricing, or positioning statements—the agent could not parse full page content from the site itself.
- ›Step [14] (Awwwards profile) and [17] (Contra freelancer profile) were the primary content-bearing sources for the founder's identity, specialization (GSAP/3D/Astro), and differentiation; these were discovered via web search, not from the site.
- ›The site's `oaksun.studio/pricing` returned a 404 (step [5]), and the agent explicitly noted in the final response: 'I could not find any pricing information on their website.' This is a critical gap for a discovery-evaluation task—the agent had to tell the user to contact them directly.
- ›The agent's final answer about 'solo founder, end-to-end ownership' and specific technical skills (Karolis Butvinskas, 15+ years, architecture background) came from external search results and Contra profile, not from the site's own content.
- ›The site is not agent-ready for this task: it does not publish structured data (schema, JSON-LD), no pricing page exists, and the direct HTML fetches were too compressed to yield semantic detail. The agent succeeded by heavy use of web search and prior knowledge (50% prior knowledge, 30% web search per metadata), not by reading the site.
Want to run your own?
Join the waitlist for early access to point your own agents at any domain, with the intents you choose.
or talk to us about agent readiness →