ora
knock.app
success
Claude Code
intent

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

12steps
61.7sduration
$0.6438cost
157,921tokens
21 steps9 reasoning steps3 searches
home
docs
docs
/product
/features
/workflows
/en-us
llms.txt
search
/knock-braze-comparison
search
search
89%
on-site discovery
67%
reliability
56%
link following
path origin
  • previous resource56%
  • web search11%
  • prior knowledge33%
insight

The agent successfully gathered a comprehensive overview of Knock's positioning, features, target audience, and pricing model through direct site exploration and comparison research. The site exposed core messaging through its homepage, llms.txt, pricing page, and Braze comparison page, but some structural details (full pricing tiers, detailed feature matrices, implementation guidance) were either inaccessible or not fully loaded. The agent assembled a satisfactory explanation by combining fetched content with prior knowledge and web search results.

  • The homepage and llms.txt (steps [1] and [11]) provided clear category positioning ('customer engagement infrastructure for product and customer messaging') and feature names, but required the agent to infer the target audience from context—no explicit 'who it's for' section was accessible.
  • Pricing was partially accessible: step [3] loaded the pricing page but the agent notes the HTML 'wouldn't load completely,' so it extracted usage-based tiers ($250/month Starter tier, free tier) from snippets or had to rely on search result metadata (G2 link) rather than complete on-page content.
  • The Braze comparison page (step [15]) was the primary source for competitive differentiation, explicitly surfaced by Knock's own site structure—this was citation-heavy in the final answer and directly answered the 'how it's different' question.
  • The site lacks dedicated 'Features,' 'Product,' or 'Alternatives' pages (steps [4], [5], [8] all returned 404s), forcing the agent to rely on homepage navigation, llms.txt summaries, and blog comparison articles rather than structured product documentation.
  • Agent-readiness was mixed: llms.txt served as an excellent machine-readable overview, but critical details (full tier comparison, Agents feature clarity, implementation time, customer count) required either search fallbacks or remained gaps—indicating the site prioritizes marketing messaging over comprehensive reference content.

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 →