Open sourcing the tooling: claude-rank, Ultraship and codesight
Three Claude Code plugins, published free on npm under the HouseofMVPs publisher account, run the same kind of technical checks we use when we review a site.
The engineering behind our technical review is public, as three Claude Code plugins on npm under the HouseofMVPs publisher account. All three are free and MIT licensed.
claude-rank is the scanner. It checks whether content is in the server rendered HTML, whether crawlers are blocked, whether structured data is valid and connected, and whether key facts are stated in visible text. Available at npmjs.com/package/@houseofmvps/claude-rank and github.com/Houseofmvps/claude-rank. The unscoped alias claude-rank-seo also resolves.
Ultraship is the builder workflow: code review, scanning, Lighthouse performance audits, security header checks, search engine submission, Playwright end to end tests and project memory. Available at npmjs.com/package/ultraship and github.com/Houseofmvps/ultraship.
codesight is a command line tool that gives AI coding agents a compact map of a codebase so they spend fewer tokens finding their way around. Available at npmjs.com/package/codesight and github.com/Houseofmvps/codesight.
A technical scan answers one question: can AI apps read the site? It does not tell you whether they name you. That takes repeated runs of real buyer questions on each AI app, which is what our baseline measures. If the scan shows a small gap, you have the recipe and can fix it yourself.