aikias.no
success
Claude Code · Haiku 4.51:28
intent
What does aikias.no do and who is it for? Explain it back to me.
16steps
88.3sduration
$1.1449cost
228,113tokens
28 steps12 reasoning steps8 searches
home
/en
search
/en/about
/en/services/automasjon
/en/ai-byra-norge
search
search
/en/faq
search
/en/tools
llms.txt
search
search
search
search
37%
on-site discovery
100%
reliability
25%
link following
path origin
- previous resource25%
- web search63%
- prior knowledge13%
insight
The agent successfully assembled a comprehensive overview of AIKI by combining web search results with a machine-readable markdown export (llms.txt). The site itself is heavily client-side rendered and largely inaccessible via direct HTML fetch, forcing reliance on search snippets and the markdown artifact. Despite this friction, the agent retrieved enough concrete detail (pricing, services, positioning, target audience) to explain the company accurately.
- ›Step [24] (llms.txt fetch) was the single most valuable source — it returned structured, human-readable content describing AIKI's services, team, founding, and pricing in markdown form. This artifact dramatically reduced the need for inference and provided the foundation for the final answer.
- ›Steps [5], [11], [12], [17], [20] (web searches) surfaced real URLs and snippets from the site that confirmed and filled gaps: pricing tiers, service categories, tools used (n8n, Make), and comparative positioning language.
- ›Direct HTML fetches of aikias.no pages ([1], [3], [7], [8], [9], [14], [22]) returned only truncated boilerplate HTML (Next.js client-side rendering), providing almost no actionable content. The agent had to work around this by relying 63% on web search and 25% on the prior-knowledge markdown artifact.
- ›The agent cited specific sources in its final response but none of the HTML fetches returned usable page bodies — all substantive content came from search results and the llms.txt export, indicating the site does not publish machine-readable, fetch-friendly versions of key pages.
- ›Pricing was one of the most concrete findings, likely sourced from llms.txt and search snippet text; service tiers (AI Partner, AI Partner Skalering, etc.) were explicitly stated with NOK amounts.
- ›The agent correctly flagged gaps: no client testimonials/case studies found, unclear service tier decision logic, no typical project cost guidance, and limited visibility into concrete use-case examples. These gaps reflect real content gaps on the published site, not just fetch failures.
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 →