data-api.md

description	Endpoint catalogue of the public content delivery API: contents, blocks, data sources, redirects, search, icons.

Data API

The public, read-only content delivery API. Base path: /api/v1. Every request authenticates with a space access token as a query parameter: ?token=… (details).

The parameter reference for filtering, sorting, language selection, and pagination lives in Querying content. This page is the endpoint catalogue.

Content

Endpoint	Description
GET /api/v1/contents	List entries. Filters: id, parent_id, canonical_id, canonical_parent_id, content_type, language, include_fallback, timestamp filters. vid=published|draft, sort, page/per_page (max 500), take/except. Revision-stamped caching.
GET /api/v1/contents/{slug}	Single entry by full slug (slashes allowed). vid, language. Localized slugs resolve when language is set.
GET /api/v1/breadcrumbs/{slug}	Ancestor trail of an entry, root first. Addressed by full slug or by content id. vid, language, include_self, ancestors, translations, include_content, take/except.
GET /api/v1/search	Full-text search over published content. q (required), language, limit, offset. Returns scored results with totals.
GET /api/v1/sitemap	Published slugs + timestamps for sitemap generation, honoring the space's sitemap-extraction settings.
GET /api/v1/sitemaps/{sitemap}	A named sitemap defined in the space's sitemaps settings — e.g. separate per-type sitemaps for pages and news. Unknown slugs return 404.

Content entry shape

{
  "id": "01JW…",
  "name": "Home",
  "slug": "home",
  "full_slug": "home",
  "language_iso": "en",
  "block": "page",
  "content": { "…fields…": "…", "nested": [{ "id": "…", "block": "teaser", "…": "…" }] },
  "position": 0,
  "published_at": "2026-07-12T09:30:00Z",
  "first_published_at": "…",
  "created_at": "…",
  "updated_at": "…"
}

content contains the field values including the full nested block tree; referenced entries, assets, and links are resolved into the payload. Per-entry cache hints (TTL, tags) configured in the entry's Config tab are delivered as response headers.

Breadcrumbs

GET /api/v1/breadcrumbs/{slug} returns the trail from the tree root down to the entry, one object per level:

{
  "breadcrumb": [
    {
      "id": "01JW…",
      "name": "Produkte",
      "slug": "produkte",
      "full_slug": "/startseite/produkte",
      "path": "/de/startseite/produkte",
      "block": "category",
      "depth": 1,
      "is_root": false,
      "is_current": false,
      "language_iso": "de",
      "resolved_language_iso": "de",
      "is_fallback": false,
      "is_published": true,
      "published_at": "…",
      "updated_at": "…"
    }
  ],
  "meta": {
    "language_iso": "de",
    "fallback_language_iso": "en",
    "i18n_mode": "overlay",
    "levels": 3,
    "root_id": "01JW…",
    "current_id": "01JX…"
  },
  "rv": 1753689600
}

Every level is resolved through its own i18n family along the requested language's fallback chain, so a level that has no translation is served from the fallback and flagged with is_fallback rather than dropped. full_slug is the stored, locale-neutral path; path is the delivery URL with the requested language's locale segment applied, even on a level that fell back.

Ancestors that are not published are omitted from the trail — an unreleased entry must not leak its name or slug through a child's breadcrumb. depth is the position in the content tree, so a gap in the sequence shows where a level was skipped. Pass ancestors=all to include them (they carry is_published: false); useful when structural folders are never published by design.

Parameter	Default	Description
language / language_iso	space default	Language every level is resolved for. Unknown values fall back to the default.
vid	published	published or draft. A version id is not accepted — every level is a different entry.
include_self	true	Include the requested entry as the last level.
ancestors	published	all includes unpublished ancestors.
translations	false	Add published sibling translations (language_iso, name, full_slug, path) per level.
include_content	false	Add the overlay-resolved content payload per level; honors take/except. Costs the version reads a lean trail avoids.

The trail costs three queries regardless of tree depth: the entry lookup, one recursive query for the ancestor chain, and one for the i18n families of all levels. include_content adds a batched resolution pass over all levels at once. Cache headers follow the requested entry's per-entry TTL and tags, exactly like contents/{slug}.

Blocks

Endpoint	Description
GET /api/v1/blocks	List block definitions (schema, type, tags). Filterable (e.g. is_nestable, tags, timestamps).
GET /api/v1/blocks/{block}	Single block definition by ID.

Data sources

Endpoint	Description
GET /api/v1/datasources	List data sources (only those marked API-available).
GET /api/v1/datasources/{slug}/entries	Entries of one source; supports dimension selection and pagination. Per-source cache TTL applies.

Redirects

Endpoint	Description
GET /api/v1/redirects	The redirect rule list (paginated; SDKs build a source → {target, status_code} map).
POST /api/v1/redirects/lookup	Resolve one source path. Body: { "source": "/old-path" }. Returns the rule or false.

Space

Endpoint	Description
GET /api/v1/spaces/me	Metadata of the token's space — including the current content revision (rv) and enabled languages.

Icons (Iconify-compatible)

Endpoint	Description
GET /api/v1/iconify/collections	Available collections
GET /api/v1/iconify/{prefix}.json	Icon data in Iconify JSON
GET /api/v1/iconify/{prefix}.svg / .css	Sprite / CSS for a collection
GET /api/v1/iconify/{prefix}/{name}.svg / .css	Single icon
GET /api/v1/iconify/search?query=…	Search icons
GET /api/v1/iconify/last-modified	Cache validation

Caching summary

Content endpoints redirect to revision-stamped URLs and are cached long; list/lookup endpoints (blocks, data sources, redirects, search, icons) use ~60-second cache lifetimes. Full mechanics: Access tokens & caching.