Skip to main content

LitPage API overview

Understand the boundary between published content delivery, content management, and authorized previews before integrating a renderer with LitPage.

Choose the right interface

LitPage separates editing a website from reading its published content. A renderer needs the published page, its public layout, language information, and SEO metadata. It does not need access to account settings or unpublished revisions.

InterfacePurposeAccess boundary
Content deliveryRead published website contentPublic published data; no content-management key
Content managementCreate, edit, organize, and publish contentAuthorized management credentials and site permissions
PreviewInspect a specific unpublished revisionA separate preview grant with a limited scope
FormsSubmit visitor informationA separate validated submission interface

The first delivery v1 resource is a published lookup for section-based pages, such as a homepage, pricing page, or campaign landing page. Collection entrypoints such as a blog index use a different content model and return 422 UNSUPPORTED_PAGE_KIND from this operation. Blog collections, document navigation, and other existing Render endpoints retain their current interfaces until their delivery contracts are documented. Do not assume that an endpoint exists by changing a resource name in the URL.

Check your deployment

This reference accompanies the delivery v1 implementation. First check that your deployment exposes:

GET /api/delivery/v1/openapi.json

The OpenAPI document lists the supported operations, request parameters, response schemas, and error responses for that deployment. A 404 at this address can mean the deployment predates delivery v1. Do not interpret an older API's response format as the new contract.

For LitPage's hosted service, use https://build.lit.page as the API origin and open the delivery OpenAPI document. For a separate deployment, use its verified API origin instead.

Read published content

A page lookup selects a registered website hostname, a page path, and an optional language. The result contains page identity and its published revision, ordered content blocks, public header and footer, appearance settings, localization data, and SEO metadata.

Saving a draft does not change the public response. Publishing a new version makes that version eligible for delivery. Header and footer publication can change the page response even when the page's own revision remains the same.

An unavailable or withdrawn page is not a successful empty page. Check the HTTP status before parsing a successful resource. Likewise, a moved page returns a typed redirect resource so a renderer can issue the website's actual 301 response.

Keep version meanings separate

  • The URL namespace delivery/v1 identifies the API compatibility boundary.
  • schemaVersion identifies the resource's data format.
  • page.revisionId identifies the published page version.
  • ETag identifies the complete selected response, including layout and SEO data.

A renderer can replace its visual components without changing authored content. Ordered blocks remain part of the page model; a React component import path or a framework-specific loader result is not part of the delivery protocol.

Next steps

Follow Read a published page for a request and a server-side JavaScript example. Use Errors, caching, and compatibility when handling failures and conditional requests. Keep editing workflows on the separately authorized management interface.