viollet.app
partial
Claude Code
intent

What does viollet.app do and who is it for? Explain it back to me.

16steps
77.4sduration
$1.2403cost
235,061tokens
27 steps11 reasoning steps8 searches
home
/en
docs
/en/pricing
/en/pricing
/features
search
search
search
search
search
search
search
docs
search
docs
100%
on-site discovery
75%
reliability
63%
link following
path origin
  • previous resource63%
  • prior knowledge38%
insight

The agent gathered a partial understanding of Viollet by combining a fetched homepage [1], a web search result snippet [20], and supplementary web searches [8]. It successfully identified what Viollet does (email-based expense tracking), who it's for (privacy-conscious Latin American users), and how it differs from competitors, but could not retrieve pricing details because the site is heavily client-side rendered and does not expose pricing information in static HTML or meta tags. The site's architecture made critical billing information inaccessible through standard fetching.

  • ›Step [1] returned the homepage HTML but it was JavaScript-heavy and did not contain readable content; the agent extracted no substantive information from the actual fetch response itself.
  • ›Step [20]'s search result snippet ('Viollet · Automatically Track Expenses from Your Email') provided the core product positioning and became the foundation for the 'What It Does' section.
  • ›Step [8] confirmed the product exists and is an expense tracker but did not return specific information about it; search results pointed to other similar products instead of Viollet-specific details.
  • ›The agent relied heavily on prior knowledge and inferences (63% of sourcing per metadata) rather than site-furnished content; it synthesized a coherent narrative but could not locate pricing, feature tiers, or transaction limits that the site presumably contains.
  • ›Multiple attempts to fetch /pricing, /en/pricing, /features, and /about all returned truncated or error responses (500 errors for /about, /docs), confirming the site does not serve meaningful content via server-side rendering.
  • ›The agent discovered no explicit competitor comparisons or feature matrices on the Viollet site itself; the YNAB/Monarch Money comparisons in the final response were constructed from prior knowledge and general fintech category understanding, not from site content.
  • ›The site's language handling (responses marked lang='es' regardless of /en path) suggests weak internationalization or a misconfigured language switcher, adding friction to discovery.

Help the next agent finish on viollet.app

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 →