krul.work
success
Claude Code
intent

What does krul.work do and who is it for? Explain it back to me.

16steps
57.8sduration
$0.6876cost
147,808tokens
26 steps10 reasoning steps3 searches
home
/en
/privacy
/en/privacy
/cookies
search
search
search
/help
docs
/en/pricing
/faq
/en/faq
docs
docs
docs
100%
on-site discovery
38%
reliability
46%
link following
path origin
  • previous resource46%
  • prior knowledge54%
insight

The agent gathered a partial but incomplete understanding of krul.work by reading the homepage and meta descriptions in both Dutch and English. It successfully identified the core service (AI-powered job application writing), target audience (Dutch-speaking job seekers), and key differentiator (personalized, honest, voice-preserving writing). However, it could not locate or confirm pricing, plan tiers, or availability details — critical information the user asked for — and had to resort to fragmentary inferences from hidden meta text and privacy policy language. The site is minimally agent-friendly: it publishes almost no structured product information beyond the homepage, hides pricing entirely, and offers no dedicated pages for FAQs, pricing, about, or help.

  • ›Steps [1] and [3] (Dutch and English homepages) contained the only substantive product information: meta descriptions confirmed the service's core function ('tailor your CV to every job you apply for'), workflow ('add a job posting and Krul tells you honestly where you stand'), and ethical positioning ('without making anything up'). No other pages on the site yielded usable content.
  • ›The agent discovered that pricing exists only indirectly, via a hidden meta tag on a 404 page mentioning 'three a month, no more' — not from published pricing documentation. This forced the agent to assemble an answer from fragments (meta tags, privacy policy mentions of 'slots') rather than from a dedicated, machine-readable resource.
  • ›The site offers no dedicated discovery pages: /pricing, /faq, /about, /blog, /docs, and /help all return 404s. The homepage itself does not link to pricing, plans, or competitive positioning. This design either reflects an early-stage product intentionally gatekeeping details, or poor information architecture. Web search also yielded no independent reviews or pricing comparisons.
  • ›The agent correctly identified that krul.work is in beta/waitlist mode, but this inference came from paraphrased text ('we're still building,' 'entre nous') rather than explicit product status statements. No clear call-to-action or sign-up flow was retrievable.
  • ›The privacy policy and cookies pages (steps [13], [15]) were reachable and well-formed but added no clarity on pricing or plans — they only reinforced the existence of a 'monthly slots' system without defining cost or limits.

Help the next agent finish on krul.work

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 →