homi.so
success
Claude Code
intent

What does homi.so do and who is it for? Explain it back to me.

9steps
54.0sduration
$0.5017cost
91,517tokens
16 steps7 reasoning steps2 searches
home
docs
docs
llms.txt
/product
search
search
docs
docs
86%
on-site discovery
100%
reliability
71%
link following
path origin
  • previous resource71%
  • web search14%
  • prior knowledge14%
insight

The agent successfully assembled a comprehensive explanation of Homi's core offering, target audience, pricing model, and differentiation by combining content from the homepage, llms.txt (which provided the richest structured narrative), and blog/changelog pages. However, the site's heavy client-side rendering made detailed feature breakdowns inaccessible—the agent had to work around incomplete HTML to deliver partial answers on pricing tier specifics and exact feature boundaries.

  • ›Step [6] (llms.txt) was the most information-dense resource, providing the core narrative about Homi's AI-story approach, multi-country support, and stage-based workflow. This file is explicitly designed for AI agent comprehension and delivered structured, quotable content the agent cited directly.
  • ›Steps [1], [3], [4] returned Next.js client-side rendered pages with minimal text content in the initial HTML—the agent could not extract concrete pricing details or feature lists from these fetches despite them returning 200 status. This forced reliance on the llms.txt artifact.
  • ›The agent supplemented missing details by inferring pricing structure ('Free → Scout → Move progression') and alternatives ('compared to Homes AI, Homebot') from partial content and prior knowledge, then explicitly flagged these gaps in the confusions section, showing transparency about what wasn't found rather than hallucinating specifics.
  • ›Step [13] (blog) and [14] (changelog) returned similarly stripped HTML and provided no additional content-bearing data, indicating the site's entire information architecture relies on client-side rendering—a poor fit for agent-native discovery.
  • ›The agent successfully identified the core differentiator (AI story + collaborative decision-making vs. filter-only tools) but acknowledged inability to access detailed comparison tables or real-world workflow examples.

Help the next agent finish on homi.so

Front builds the agent version of your site and shows which agents visit, what they came for, and what your pages couldn't answer.

or talk to us about agent readiness →