kapso.com
success
Claude Code
intent
What does kapso.com do and who is it for? Explain it back to me.
16steps
60.4sduration
$0.7089cost
178,824tokens
25 steps9 reasoning steps4 searches
home
docs
docs
/features
docs
docs
/login
search
/platform
/item
/alternatives/kapso
search
/twilio-alternative-fo…
search
search
/dashboard
67%
on-site discovery
42%
reliability
8%
link following
path origin
- previous resource8%
- web search33%
- prior knowledge58%
insight
The agent successfully understood Kapso's core offering, pricing, and differentiation by fetching the homepage, pricing page, platform page, and Twilio comparison page. The site's public pages were navigable and content-rich, but incomplete—multiple standard pages returned 404s, and the agent had to rely on web search and prior knowledge (58% of fetches) to fill gaps and cross-reference information from external sources like Hacker News and archyde.com.
- ›Step [1] (homepage) established Kapso's core pitch ('WhatsApp for developers') and basic positioning in metadata. Step [2] (pricing) provided concrete pricing tiers (Free, Pro $25, Platform $299). Step [13] (platform page) clarified the multi-tenant and automation layer. Step [21] (Twilio comparison) was explicitly cited and directly compared pricing and features.
- ›The site publishes its key selling points on the homepage and pricing page in well-structured HTML, but the agent had to use web search (steps [11], [18], [19], [23]) to discover the Twilio comparison, platform page URL, and founder/use-case context—these were not linked from the homepage's navigation.
- ›Site incompleteness created friction: /docs, /features, /about, /login, /dashboard all returned 404s. The agent worked around this by discovering /platform and /twilio-alternative-for-whatsapp via search results, and by extracting founder insights from Hacker News and third-party coverage. The site is not fully self-contained for understanding the product.
- ›Agent-readiness is moderate: pricing and positioning are clear and machine-readable, but the site lacks structured comparison data, published use-case documentation, and a complete information architecture. The agent had to synthesize information across 4 successful content fetches + multiple web searches to produce a comprehensive answer.
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 →