cbthera.com
success
Claude Code
intent
What does cbthera.com do and who is it for? Explain it back to me.
7steps
29.7sduration
$0.0369cost
18,901tokens
9 steps2 reasoning steps
home
docs
docs
/platform
/compounds
/markets
/solutions
80%
answer from your site
29%
answer efficiency
67%
followed site links
answer sources
- from this site80%
- from agent knowledge20%
url discovery
- given in the task14%
- followed a link57%
- guessed the URL29%
insight
Agent successfully synthesized a coherent overview of CB Therapeutics' business from two fetched pages (homepage and platform). The site is moderately navigable but sparse: it contains core positioning content but lacks customer information, detailed pricing, regulatory context, and typical B2B onboarding flows; the agent filled gaps using domain knowledge (e.g., inferring B2B licensing models) rather than site content.
- ›Only 2 of 7 fetches succeeded (steps 1 and 4); the rest were 404s (steps 2, 3, 5, 6, 7). This suggests the site has a minimal navigation structure with most intuitive URLs non-functional — /pricing, /about, /compounds, /markets, /solutions all returned 404, forcing reliance on homepage and /platform alone.
- ›The homepage and /platform page were content-rich and well-structured, delivering: business model (precision fermentation), target segments (pharma, cannabinoids, food, fragrance), IP portfolio (8 patents), and a clear competitive advantage table vs. traditional agriculture. No truncation in either fetch indicates full content was retrieved.
- ›Pricing is genuinely absent from the site; the agent correctly identified this as a gap rather than guessing, and inferred from the language 'collaboration model' and 'licensing arrangements' that pricing is negotiated B2B rather than published. This inference was grounded in site content, not model memory.
- ›Missing content the agent flagged (case studies, team, funding, regulatory compliance, partnership initiation process) is plausible for a precision biotech company but was never surfaced as fetchable URLs by the site itself. The agent tried conventional paths (/about, /compounds, /solutions) without success, suggesting either poor information architecture or intentional omission.
- ›The agent did not discover these resources via search or external sources; 43% of URLs were typed from memory (intuitive guesses like /pricing), and 57% followed links from fetched pages. This indicates the site controls its own discoverability, and what is not linked or published on the two successful pages is effectively unreachable.
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 →