Skip to content
BetaDocs

Build a blog with Pages

This is the tr-blog recipe exactly as the Typeroll MCP server gives it to agents through list_skills and read_skill. See Skills and recipes for how agents load recipes.

Every article is a Page. A Content type supplies its extra fields, URL pattern and default template. Do not create a separate item store or generate listing HTML by hand.

  1. Read get_site_capabilities, list_block_types, list_content_types and list_page_templates. For a substantial change, create a site version and pass its ID as version on every supported operation.
  2. Create an editable shared template:
{ "name": "article-layout", "label": "Article", "starter": "article", "status": "published" }

Send it to create_page_template. Use block tools with target: { kind: "template", id: "article-layout" } to refine the layout.

  1. Call create_content_type:
{
  "name": "articles", "label_singular": "Article", "label_plural": "Articles",
  "route_template": "/articles/{slug}", "template": "article-layout",
  "sort_field": "date_published", "sort_dir": "desc",
  "fields": [
    { "name": "excerpt", "label": "Summary", "type": "textarea" },
    { "name": "featured_image", "label": "Cover image", "type": "image" }
  ]
}

Title, slug, author, date, SEO, status and body are built-in Page values. Keep only custom values in fields.

  1. Call create_page for each article with its title, slug, content_type: "articles", status, date_published, fields and blocks. Use native heading, prose, image, list and table blocks. Store its returned Page ID; use it for every later operation.
  2. Create the archive as another Page with core/page_list, using data: { content_type: "articles", sort_by: "date_published", sort_order: "desc", paginate: 10 }. A homepage teaser can use limit: 6 instead. Both resolve saved Pages at render time; adding an article needs no listing regeneration.
  3. Link to the archive from navigation. Preview representative articles and the archive on desktop and mobile with get_preview_link page_id=....
  4. Save any working copies before publication. update_page and block edits need save: true or commit_working_copy; new Pages and template definitions are saved immediately. Deploy only within the user’s authorized scope and verify the returned job and public URL.

Use the organization’s connected media storage before importing remote images. Do not hardcode a Typeroll media hostname. Use the returned media URL.

A published Page with complete route tokens becomes one static detail page. Draft/review Pages stay private. Unlisted Pages have a public URL but are omitted from listings and search indexes. The selected version’s robots policy applies.