Murit CMS Developer Documentation

Returns the public metadata and configuration for the specified site.

{
  "data": {
    "siteKey": "site_demo_acme",
    "name": "Acme Tech Blog",
    "defaultLocale": "en",
    "primaryURL": "https://acme.com",
    "blogBasePath": "/blogs/",
    "timezone": "UTC",
    "seoDefaults": {}
  }
}

Returns a paginated list of published post summaries. Does not include full body HTML for optimal payload size.

{
  "data": [
    {
      "id": "123",
      "title": "Decoupling Publishing from Deployments",
      "slug": "decoupling-publishing-from-deployments",
      "excerpt": "How Murit CMS decouples content publishing from website deploys.",
      "publishedAt": "2026-09-06T12:00:00.000Z",
      "modifiedAt": "2026-09-06T12:00:00.000Z",
      "featuredMedia": {
        "url": "https://cdn.domain.com/media/article-cover.webp",
        "alt": "System diagram",
        "caption": null,
        "width": 1200,
        "height": 630
      },
      "categories": [{ "name": "Engineering", "slug": "engineering" }],
      "tags": [{ "name": "Architecture", "slug": "architecture" }],
      "seo": {
        "title": "Decoupling Publishing from Deployments",
        "description": "How Murit CMS decouples content publishing from website deploys.",
        "canonicalURL": "https://domain.com/blog/decoupling-publishing-from-deployments/",
        "keywords": null,
        "socialImage": {
          "url": "https://cdn.domain.com/media/article-cover.webp",
          "alt": "System diagram",
          "caption": null,
          "width": 1200,
          "height": 630
        }
      }
    }
  ],
  "meta": {
    "limit": 20,
    "nextCursor": "eyJpZCI6MTIzfQ=="
  }
}

Returns a single published article by slug with full metadata and sanitized contentHtml. If the requested path matches a published post's legacyUrl or redirectUrls entry (absolute permalinks, root-relative paths, and multi-segment paths all resolve), the API answers with an HTTP 308 Permanent Redirect whose Location is the canonical public site URL, plus destination slug metadata in the body.

{
  "data": {
    "id": "123",
    "title": "Decoupling Publishing from Deployments",
    "slug": "decoupling-publishing-from-deployments",
    "excerpt": "How Murit CMS decouples content publishing from website deploys.",
    "publishedAt": "2026-09-06T12:00:00.000Z",
    "modifiedAt": "2026-09-06T12:00:00.000Z",
    "featuredMedia": { "url": "https://cdn.domain.com/media/cover.webp", "alt": "Cover Image", "caption": null, "width": 1200, "height": 630 },
    "contentHtml": "<p>When content authors update copy, you shouldn't have to wait for CI/CD...</p><h2>How It Works</h2><p>Our Content API serves cached responses with automatic ETag revalidation.</p>",
    "categories": [{ "name": "Engineering", "slug": "engineering" }],
    "tags": [{ "name": "Architecture", "slug": "architecture" }],
    "seo": {
      "title": "Decoupling Publishing from Deployments",
      "description": "Complete breakdown of the v1 Content API contract.",
      "canonicalURL": "https://domain.com/blog/decoupling-publishing-from-deployments/",
      "keywords": null,
      "socialImage": {
        "url": "https://cdn.domain.com/media/cover.webp",
        "alt": "Cover Image",
        "caption": null,
        "width": 1200,
        "height": 630
      }
    }
  }
}

Accepts limit (1-500) and cursor. Powers your dynamic sitemap without needing to load bulky article bodies.

Redeems an ephemeral 5-minute cryptographic token issued by Murit CMS to inspect the latest saved draft before it is publicly published. Returns the full post DTO with Cache-Control: private, no-store.

Returns { posts: [...] } matching the exact Astro 5 & legacy blog schema with zero-double-slash root URLs, slug preservation (casing & hyphens retained), rich-text Lexical HTML (including YouTube & Vimeo embeds, uploaded & streamed videos, Social Embeds (X/Twitter, Instagram, LinkedIn, and Facebook with 50% compact sizing), responsive column grids, image captions and links, and sanitized .content-legacy-html blocks for imported tables, galleries, and buttons), and redirectUrls for 301 legacy mappings.

Note: You can also append ?format=astro to /posts and /posts/{slug} to receive this flat Astro DTO structure.

{
  "posts": [
    {
      "id": "ibc-supreme-court-ruling",
      "title": "IBC Supreme Court Landmark Ruling",
      "date": "2024-03-15T10:00:00.000Z",
      "excerpt": "Comprehensive legal summary of the IBC ruling.",
      "image": "https://cdn.domain.com/media/hero.png",
      "imageAlt": "Court gavel",
      "imageWidth": 1200,
      "imageHeight": 630,
      "seoImage": "https://cdn.domain.com/media/hero.png",
      "seoImageWidth": 1200,
      "seoImageHeight": 630,
      "categories": ["Corporate Law"],
      "tags": ["IBC"],
      "seoKeywords": "ibc, supreme court, insolvency",
      "seoTitle": "IBC Supreme Court Landmark Ruling",
      "seoDescription": "Comprehensive legal summary of the IBC ruling.",
      "contentHtml": "<h2>Overview</h2><p>Article content...</p>",
      "contentMarkdown": "## Overview\n\nArticle content...",
      "updatedAt": "2024-03-16T09:00:00.000Z",
      "lastUpdatedAt": "2024-03-16T09:00:00.000Z",
      "legacyUrl": "https://rjls.in/ibc-supreme-court-ruling/",
      "redirectUrls": ["https://rjls.in/ibc-supreme-court-ruling/"]
    }
  ]
}

Authenticated endpoint for automated legacy blog migration. Fully supported platforms: WordPress (block and classic editor, via the REST API), Shopify (Admin GraphQL API or public blog feed), and Payload CMS (standard Payload 3 collection endpoints and custom RJLS endpoints), plus Murit/Payload JSON exports. Other JSON feeds are imported on a best-effort basis. Available directly in the Admin Console under the Migrate Legacy Blogs tab on the dedicated project page (/admin?project=<id>&tab=migration) and from Overview. CMS operators must first deploy the registered 20260909_122610_legacy_redirects database migration, which adds legacy URL fields to posts and saved versions. Frontend clients do not run database migrations.

Best-effort migration: importing creates a copy in Murit CMS; the original post on the source site is never modified. Only markup the importer recognises reliably (structure, platform CSS classes, or schema.org metadata) becomes a native block. Everything else is preserved inside a sanitized Legacy HTML block and flagged for review rather than discarded. Detection can still be wrong: review the whole article before publishing, and if you re-edit an imported post without reviewing it, an incorrect or unusual result is possible. Keep a backup of the source site, and remember that re-publishing only changes the Murit copy. Inline images remain hotlinked until media ingestion runs.

HTML Entity Decoding & Editor Sanitization: The migration pipeline and collection hooks automatically sanitize and decode HTML entities (such as ', ", &, ’, “, ”, –) across post titles, excerpts, Lexical AST rich text nodes, taxonomies, and SEO metadata. Articles appear natural and readable in the Payload CMS editor with genuine quotes and punctuation. Existing database records can be scanned and repaired anytime via npm run repair:entities.

Native Also Read & Social Embed Conversion: Legacy links formatted as "Also Read:", "Read Also:", "Read More:", or <aside class="also-read-link"> are automatically converted into native RelatedArticle blocks and resolved against existing and newly imported website articles. Social media embeds (Twitter/X blockquotes and status URLs, Instagram, LinkedIn, and Facebook embeds) are converted into native SocialEmbed (xEmbed) blocks with server-side metadata enrichment and tracking-safe card previews, while YouTube and Vimeo embeds become native video blocks, Spotify, SoundCloud, Google Maps, TikTok and CodePen embeds become native embed blocks, and notices, alerts, tips, Ghost callout cards and Elementor icon boxes become native callout blocks. For Payload 3 migrations, block aliases (such as twitter, alsoRead, youtube, divider) and legacy upload nodes are automatically normalized to native Murit blocks, while blocks that cannot be represented safely are flattened into native Lexical elements or kept as Legacy HTML so no content is lost and the editor never crashes. This normalization also runs whenever an article is opened or saved in the editor, so supported blocks that older sources stored as plain paragraphs (including Also Read: links and bare social/YouTube/Vimeo URLs) are promoted to editable native blocks instead of remaining text.

Element-accurate HTML conversion: Rendered WordPress and Shopify HTML is parsed as a document tree, so whole elements are recognized before conversion: images (with captions, links, lazy-load and protocol-relative URLs) become externalImage blocks; YouTube, Vimeo (including unlisted privacy hashes) and MP4/WebM videos become video blocks with captions; tweets, Instagram, Facebook and LinkedIn embeds become exactly one xEmbed block each; Gutenberg columns and media & text sections become columns blocks with their widths; separators become horizontal rules; lists and quotes keep their structure; <pre> code becomes an editable code block that keeps its language and renders as <pre><code class="language-{lang}">. Inline links to YouTube stay ordinary links. Complex tables, mixed-content galleries/theme sections, prose-wrapped button groups, forms, plugin widgets and other unsupported markup are kept as a sanitized legacyHtml block with a sandboxed editor preview instead of being flattened to text. A fidelity check compares every visible word and media URL of the source with the result and reports anything that did not carry over.

Migration report: preview and import responses include report ({ articles, cleanArticles, elements: { images, videos, embeds, callouts, buttons, faqs, galleries, tables, socialEmbeds, alsoRead, columns, horizontalRules, legacyHtml }, preserved: { [label]: count }, needsReview: [{ title, slug, postId?, preserved, warnings }] }) and skippedCount (items without a title). import also returns the saved postIds; after the last batch, call { action: "resolve-links", postIds } to link Also Read blocks to articles imported in later batches (returns { linked, updatedPosts }).

Legacy Blog Migration provides a 100% self-serve, in-browser media ingestion flow and REST API. After importing articles, tenants can trigger Ingest All Media Assets into Storage with a single click directly inside the Admin Console (no CLI or server access required). Media files are downloaded and uploaded in resilient 5-item batches to prevent community proxy timeouts (Cloudflare 524, Vercel 60s), with real-time progress, pause/resume controls, and automatic 404 skip handling. In CDN mode, ingested inline images are upgraded to native upload cards while keeping their alt text, caption, link, alignment and width; relative rewrite mode keeps the legacy `/wp-content/uploads/...` paths.

For zero broken backlinks and 100% SEO rank preservation, choose Same-Origin Exact Path Preservation (relative paths like /wp-content/uploads/photo.webp). Your frontend serves them via a single-line reverse proxy rewrite in astro.config.mjs or next.config.js directly from your own domain.

CLI automation is also supported for headless scripting:

# Pre-flight audit without modifying data
npm run media:migrate -- --website <ID> --dry-run

# Ingest into connected BYOS bucket (R2/S3/GCS) with retry & idempotency
npm run media:migrate -- --website <ID> --rewrite-mode relative

Authenticated endpoint that fetches server-side metadata and link previews for social posts (LinkedIn, X/Twitter, Instagram, Facebook). Resolves author identity, text content, image assets, and canonical URLs directly on the server to prevent client-side browser tracking prevention blocks (e.g. Microsoft Edge Tracking Prevention or Brave Shields blocking third-party iframe embeds). Cached in-memory with a 10-minute TTL.

Authenticated endpoint that resolves a pasted link to a direct image URL. Direct image links pass through unchanged; web page links (e.g. an Unsplash photo page) are resolved server-side from their OpenGraph preview image, so authors can insert an external image from either a direct file URL or a page URL. Non-http(s) URLs and private/internal hosts are rejected. The image stays on its original host and is referenced by URL — it is never copied to storage.

Authenticated endpoint for requesting project deletion with a 48-hour grace period timeline. Requires project manager permissions (Platform Admin, Org Admin, or Site Admin). Confirmed by passing either the exact project name or the keyword DELETE in uppercase.

Effects: Sets website status to pending_deletion, records deletionRequestedAt and scheduledForDeletionAt (48 hours from request), and immediately pauses public Content API queries (returns 404). All articles, media, and settings remain safe and intact during the grace period. Once the deadline passes, the purge scheduler permanently deletes the website, its posts, media files (including cloud storage objects), categories, tags, and scoped memberships.

Authenticated endpoint for reversing a project deletion request within the 48-hour grace period. Requires project manager permissions.

Effects: Resets website status to active, clears deletion timestamps, cancels the scheduled purge, and restores immediate live Content API delivery. After the deadline the request is rejected with HTTP 410 and the project is permanently purged instead.

Expired projects are also removed from GET /api/workspace, so they never reappear in the console even if a purge attempt is delayed.