isaac.fyi
success
Claude Code
intent

What does isaac.fyi do and who is it for? Explain it back to me.

20steps
110.9sduration
$1.7319cost
159,630tokens
33 steps13 reasoning steps16 searches
home
home
search
search
search
search
search
search
/api/guestbook
search
home
search
search
search
search
search
search
search
search
search
75%
on-site discovery
100%
reliability
50%
link following
path origin
  • previous resource50%
  • web search25%
  • prior knowledge25%
insight

The agent successfully understood isaac.fyi by assembling information from web search (which found Isaac Ng's LinkedIn and Zora profile) and by accessing the site's guestbook API, which returned structured data listing his projects, reading list, and themes. However, the site itself did not render its core content in a machine-readable form — the agent could not fetch the homepage HTML in a usable way and had to rely on prior knowledge of Isaac Ng's work and web search to reconstruct what isaac.fyi contains. The site is not agent-ready; it appears to be a client-side rendered portfolio with no structured metadata, API documentation, or alternative access patterns for machines.

  • ›Step [15] returned the guestbook API response — the ONLY machine-readable content the agent accessed directly from isaac.fyi itself. This endpoint exposed project names, reads, and thoughts, but only as bare lists without descriptions or linking.
  • ›Steps [1] and [3] failed to retrieve usable homepage content due to dynamic/client-side rendering (JavaScript-heavy Next.js site); the agent received truncated HTML that did not contain the actual site content.
  • ›The agent relied on web search results (steps [5], [12], [29]) to identify Isaac Ng as Head of Product at Zora and to find his LinkedIn profile, which provided biographical context. This information was NOT on isaac.fyi itself — it came from external web presence.
  • ›Steps [17], [21] attempted to fetch hardtobook.xyz (one of the 'builds' mentioned in guestbook data), confirming it exists but returning only raw HTML without meaningful semantic content.
  • ›The site published NO structured metadata (schema.org, microdata, or JSON-LD), no /about or /projects endpoint with machine-readable descriptions, and no llms.txt or similar AI-consumption format. The guestbook API is the only intentional machine interface.
  • ›The agent could not verify or retrieve details for most 'builds' projects (subject, foundry, homer.quest, manufacture.wiki, daos) — these appear to be either internal, early-stage, or exist only as references on isaac.fyi with no web presence to corroborate them.

See every agent that visits isaac.fyi

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 →