plainrouter.com
success
Claude Code
intent
What does plainrouter.com do and who is it for? Explain it back to me.
22steps
84.2sduration
$1.0388cost
346,237tokens
35 steps13 reasoning steps5 searches
home
api-docs
openapi
docs
.well-known
docs
search
search
search
docs
docs
docs
docs
docs
/dashboard
/register
docs
search
search
docs
/features
docs
100%
on-site discovery
71%
reliability
76%
link following
path origin
- previous resource76%
- prior knowledge24%
insight
The agent successfully assembled a comprehensive explanation of PlainRouter by combining API documentation, markdown guides, and the OpenAPI spec—all discoverable through direct URL guesses and routing from the docs index. However, the site's agent-readiness was uneven: core identity (what it does, who it's for, how it differs) was well-documented in markdown and API specs, but pricing information was entirely absent, forcing the agent to explicitly call out a gap rather than deliver a complete answer to the task.
- ›The agent extracted the core value proposition ('first-party conversion signals platform for Meta') from three content-bearing sources: the API docs overview (step 3), the markdown index and quickstart (steps 25–26), and the Signals overview (step 27). These were discoverable only by guessing or following routing hints in llms.txt (step 23), not from homepage content.
- ›Pricing was completely absent from the site—the agent checked /pricing (step 4, which returned a client-side rendered page with no pricing content), found no public docs on it, and confirmed via web search that no pricing is published. The agent explicitly acknowledged this gap in the final response rather than speculating, meeting transparency expectations for a discovery-evaluation task.
- ›The site's agent-readiness was bifurcated: markdown documentation (auth.md, docs/*.md) and the OpenAPI spec (step 7) provided machine-readable, well-structured answers to 'what,' 'who,' and 'how different'; but the homepage and client-side pages (steps 1, 4, 32, 33) were Inertia-rendered React apps with no static content for agents to parse. The llms.txt file (step 23) was crucial—it acted as a discovery mechanism for the agent to find all available documentation.
- ›The agent had to synthesize positioning and differentiation by cross-referencing multiple markdown docs (Signals overview, quickstart, auth.md) rather than finding it pre-composed on a marketing page. This required assembly work but did yield a coherent answer grounded in real product design (first-party CNAME, MCP integration, governed AI actions).
Help the next agent finish on plainrouter.com
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 →