mxa-discoverability
Builds search and AI-answer visibility in while a site is being made: metadata and social cards per route, canonical URLs, language versions, structured data, sitemaps and feeds, and content that stays addressable and citable.
- Puts a title, description, social card and canonical on every route, written from that page's own content and sent by the server, so nothing inherits them from the root or rewrites them after the page loads.
- Keeps listings, filters and pagination at addressable URLs where the surface warrants it, so a page can be linked to and cited.
- Builds the machine-facing surfaces as part of the site: the sitemap, feeds, and markdown served on request.
- Builds the redirect map from the old sitemap before a site with existing URLs is replaced, in the framework's own config, held by a test.
- Verifies the site with Search Console and Bing Webmaster Tools at setup, so the engines' own reports are there to read later.
- Gives each language version its own address, links the versions to each other, and keeps untranslated copies out of search.
- Shows authorship, dates and the source of every figure on the page, because metadata an assistant skips does not travel.
- Uses one name for the company everywhere and adds the structured data with a documented use, claiming what each type does and never a citation.
What it uses
Framework-agnostic. It works with whatever the project is already built on.
Install it
Already on studio Claude accounts, in the Made by Agents plugin. Anyone else:
claude plugin marketplace add madebymany/madebyagentsclaude plugin install made-by-agents@madebymanyFor Cursor, Codex and other agents, just this skill:
npx skills add madebymany/madebyagents --skill mxa-discoverability