roji.ai
success
Claude Code
intent
What does roji.ai do and who is it for? Explain it back to me.
6steps
32.1sduration
$0.1714cost
47,861tokens
7 steps1 reasoning step
home
/index.md
/skill.md
llms-full.txt
docs
docs
100%
on-site discovery
100%
reliability
67%
link following
path origin
- previous resource67%
- prior knowledge33%
insight
The agent successfully gathered a comprehensive explanation of roji.ai's business model, pricing, and positioning by fetching the homepage markdown, skill documentation, and pricing page. The site publishes core information in machine-readable markdown format (index.md, pricing/index.md) alongside HTML, making it highly navigable for agents; however, some technical details (x402, AP2) and competitive context remain unexplained or undiscovered.
- ›Step [1] (index.md) and Step [2] (skill.md) together provided the core value proposition, use cases, and architectural approach (MCP + OAuth 2.1, SLA-backed ordering, instant settlement). Both were machine-readable markdown, not scraped from HTML.
- ›Step [5] (pricing/index.md) delivered complete pricing logic: 15% platform fee (paid by buyer), volume discounts (12%/10%), 100% list price to business, $0 to list—all structured and unambiguous.
- ›The site proactively publishes markdown source files (index.md, pricing/index.md, skill.md) alongside HTML, signaling strong agent-readiness; the agent accessed these via prior knowledge guesses rather than site navigation, suggesting they are not discoverable from HTML navigation alone.
- ›Significant gaps remain: x402 stablecoin payments, AP2 mandates, and service directory scope are mentioned but unexplained; no competitor comparison exists; geographic/regulatory constraints are implied but not formalized.
- ›The agent did not access an actual service catalog or browse available verticals, so practical availability and demand-side usability remain untested against the pitch.
See every agent that visits roji.ai
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 →