ora
cmu.edu
success
Claude Code
intent

What does cmu.edu do and who is it for? Explain it back to me.

15steps
56.2sduration
$0.6223cost
160,579tokens
19 steps4 reasoning steps3 searches
home
/admissions
/academics
docs
docs
docs
/schools
search
/admission/costs-aid
/student-admission/acc…
search
search
docs
/admission/who-we-are
/admissions-and-financ…
67%
on-site discovery
58%
reliability
58%
link following
path origin
  • previous resource58%
  • web search33%
  • prior knowledge8%
insight

The agent assembled a comprehensive overview of CMU by combining homepage metadata, direct navigation to /academics and /about pages, and targeted web searches for admissions and financial aid details. The site was partially navigable via its main sections but required search fallback when subpaths like /about/rankings-and-facts returned 404s; the agent successfully synthesized scattered content into a coherent explanation of what CMU does, who it serves, pricing, and differentiation, though some details (graduate costs, total enrollment) remained incomplete.

  • Step [1] (homepage) provided meta description positioning CMU as 'a private, global research university' with 'cutting-edge programs,' establishing foundational identity; steps [3] and [4] (/academics and /about) gave breadth by exposing school structure and mission language ('pushes boundaries of innovation').
  • Steps [11–12] (admission/costs-aid and access-affordability pages) contained explicit pricing data ($969 per unit, $75k–$100k aid thresholds, 100% need-met policy) that directly answered the 'how it's priced' question, discovered via web search rather than site navigation (direct /admissions URL returned 404 at step [2]).
  • Step [16–17] (vision-mission-values and who-we-are pages) sourced the 'That's So CMU' brand positioning and interdisciplinary ethos; these were discovered through targeted search for 'distinctive strengths,' indicating the site does not surface differentiation prominently via homepage navigation.
  • The site's information architecture is fragmented: admissions lives at cmu.edu/admission (not /admissions), course catalog at coursecatalog.web.cmu.edu, and some /about subpaths return 404s despite being structurally guessable, forcing the agent to rely on web search (33% of fetches) and prior knowledge (8%) to fill gaps that direct navigation could not resolve.
  • The agent could not locate graduate enrollment numbers, total student population, or complete graduate financial aid details on the fetched pages, and the site does not explicitly compare itself to competitors (MIT, Stanford, etc.), limiting the comprehensiveness of the 'how it's different' section.

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 →