The schema.org markup emitted on every page so AI citation engines can extract claims cleanly.
Every public page on GM Markets emits structured data as JSON-LD in the document head. This is the same data /llms-full.txt exposes as plain text, plus a layer of schema typing that AI citation engines and search crawlers consume directly.
| Schema | Where it appears | What it carries |
|---|---|---|
Organization | Every page (root layout) | Brand identity, logo, contact point, social profiles |
WebSite | Every page (root layout) | Site identity, search action |
BreadcrumbList | Asset detail, docs, list pages | Hierarchical position in the site |
FAQPage | /docs/faq, key asset pages | Question / answer pairs extractable by AI agents |
Article | Docs pages | Page-level metadata: dateModified, author |
FinancialProduct | Asset detail (/asset/<ticker>) | Asset-level metadata for the tokenized instrument |
JSON-LD is the format Google, Perplexity, and Claude crawlers extract most reliably. It does not interfere with rendered HTML, sits at the document head, and is preserved through aggressive transforms like reader-mode and AMP variants.
Open the source of any GM Markets page and search for application/ld+json. Every block is a single, well-formed JSON document.
Beyond schema markup, GM Markets ships:
gm-markets-mcp) for Claude Desktop and other MCP-aware clientsThese turn GM Markets into a tool agents can call, not just a document they cite.