upstream.do
success
Claude Code
intent
What does upstream.do do and who is it for? Explain it back to me.
6steps
38.5sduration
$0.0430cost
25,376tokens
8 steps2 reasoning steps
home
docs
/features
docs
home
docs
100%
answer from your site
50%
answer efficiency
80%
followed site links
answer sources
- from this site100%
url discovery
- given in the task17%
- followed a link67%
- guessed the URL17%
insight
The agent retrieved core product, pricing, and audience information from three successful page fetches and assembled a coherent explanation of Upstream's value proposition. However, the site itself provides minimal differentiation details, no competitor context, and vague positioning on key features like the 'Agents' use case, forcing the agent to acknowledge significant gaps rather than answer the full question.
- ›Step [1] (homepage) and step [4] (homepage re-fetch) both returned the same high-level product overview and use case framing—the agent had to fetch the homepage twice, suggesting the site structure relies heavily on a single entry point with limited internal navigation.
- ›Step [2] (pricing page) was the only dedicated feature page successfully fetched; steps [3], [5], and [6] (about, features, blog) all returned 404s, indicating a minimal information architecture that lacks standard support pages.
- ›The agent correctly identified that the site provides no explicit competitor comparisons or positioning against alternatives—this is a genuine site limitation, not model memory, confirmed by the agent's statement 'The webpage contains no explicit mentions of competitors.'
- ›The 'Agents' positioning appears in the site content (step [4] references 'For Agents' section) but lacks explanation—the agent noted this as confusing rather than filling the gap with external knowledge, appropriately grounding its uncertainty in what was actually surfaced.
- ›Pricing information was complete and well-structured on the dedicated pricing page (step [2]), making that section of the final response fully site-grounded; by contrast, differentiation and use-case details required the agent to acknowledge vagueness and gaps rather than speculate.
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 →