llms.txt served on your domain
Discovery file
GET /api/content (no auth required)
Content feed
POST /api/qualify/{orgSlug}
Hand-off endpoint
Agents interact with Spectare in three sequential steps. First, the entry point: your site serves an llms.txt file that maps what you publish and where each atom lives as clean JSON. An agent that reads your site gets your facts accurately from a handful of fetches rather than scraping rendered HTML across dozens of pages. The file is generated from your live library and updates as you publish.
Second, content: every published atom is available at GET /api/content and GET /api/content/{slug}, each returning structured JSON with a stable slug. An agent can fetch the index or jump directly to a specific atom. No credentials are required for public content. The same endpoints are available on your own domain when you use the WordPress plugin or the SSR helper.
Third, hand-off: an agent can POST to /api/qualify/{orgSlug} with a plain-text summary of what its user is looking for and receive back a personalised page URL. That link carries a context token so when the visitor clicks through, Spectare assembles a page built for that specific intent rather than a generic one. The MCP server exposes the qualify endpoint as a native tool, so any MCP-compatible assistant can call it without custom integration work.
This is one atom from the Spectare content library. Spectare assembles the right atoms for each visitor in real time, based on who they are and how they arrived.