What does callumgordon.dev do and who is it for? Explain it back to me.
- previous resource67%
- prior knowledge33%
The agent successfully understood callumgordon.dev as a personal technical portfolio and blog, not a commercial product. It assembled the answer by combining prior knowledge about Callum Gordon's professional background (Typeform, MCP expertise) with fragments fetched from the site itself: the homepage and writing section provided limited direct content, but the RSS feed and article fetch confirmed the site's focus on MCP tool design and AI evals. The site was minimally navigable for this task—compressed HTML responses prevented rendering key homepage details, forcing the agent to rely on guesswork and prior knowledge rather than discoverable, machine-readable content.
- ›The homepage (step [1]) returned Brotli-compressed HTML that the agent could not fully parse; no structured metadata, FAQ, or about section was extracted, leaving the agent unable to confirm key claims from primary content.
- ›The RSS feed (step [17]) was the only machine-readable, content-bearing artifact the agent obtained, confirming the blog's focus on 'MCP tool design, evals and automation' via the feed description, and yielding one article title.
- ›The agent cited three URLs in its Sources section ([callumgordon.dev/](https://callumgordon.dev/), [callumgordon.dev/writing](https://callumgordon.dev/writing), and a LinkedIn profile), but only two were successfully fetched with usable content (homepage and writing page); the /about endpoint returned 404, and the homepage content remained opaque.
- ›The agent filled most of its final answer (Callum's role at Typeform, technical skills, location, GitHub profile) using prior knowledge (33% of fetches sourced from prior knowledge), not content discovered on the site, indicating the site does not surface basic identity information in an agent-ready format.
- ›The site's navigation is minimal and unaggressive: the agent had to guess at /about, /projects, and other common portfolio routes (many returned 404); the RSS feed was not linked from fetched pages and was only discovered via a second fetch attempt.
Help the next agent finish on callumgordon.dev
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 →