What Lighthouse measures
Lighthouse produces lab evidence about one rendered page: performance metrics, accessibility violations, best-practice checks, and basic SEO signals, all from a controlled browser run against a single load. It answers a narrow question well: is this page fast, accessible, and built according to web platform guidance. It is the reference tool for that question, and Santos uses it rather than reinventing it.
What traditional SEO tools measure
Platforms such as Ahrefs and Semrush, and generic site crawlers, measure search discovery and ranking: crawl coverage, indexation, keyword positions, backlink profiles, SERP features, and competitive gaps. They answer whether search engines can find and rank your content, and they do it at a scale of historical data that a point-in-time audit cannot match.
What Santos measures
Santos measures agent readiness across four dimensions: whether an autonomous system can discover the site, understand what it offers, invoke applicable capabilities, and rely on the interaction. That includes surfaces the other tools do not model at all — OpenAPI documents, MCP servers, capability manifests, typed error behavior, and x402 payment challenges — scored deterministically and adjusted for applicability.
Where they overlap
Performance and accessibility appear in both Lighthouse output and the Santos Trustworthy dimension. Crawlability, canonical URLs, sitemaps, and metadata quality matter to both SEO tools and the Discoverable dimension. This overlap is an asset: fixing the foundational layer improves all three families of scores at once, which is why the cheapest readiness work is usually the oldest work.
Where they do not overlap
A perfect Lighthouse score says nothing about how an agent would call your API, because invocation surfaces are outside its model. A top-tier SEO audit will not check whether you publish MCP tools, a capability manifest, typed schemas, or a well-formed 402 challenge, because those signals do not influence search ranking. The gap is not a flaw in either tool; it is simply a different question.
When to use which, and how they compose
Use Lighthouse when the question is page quality, SEO platforms when the question is search growth, and Santos when the question is whether machines can act on your site without a human in the loop. The tools compose rather than compete: Santos Deep audits run Lighthouse and axe-core as one evidence layer among others, then add interface discovery, commerce, and consistency evidence on top. A mature team runs all three and reads each score for the question it actually answers.