Content Production & Publishing

What this area provides

The named capabilities in Content Production & Publishing. Each is referenced from the workflows that use it.

• Content Creation

  AlmaSEO's Content Creation engine lets agencies and freelancers generate publish-ready SEO articles from a single form. Users choose from 11 specialized article types (blog post, service page, city/county/state landing page, FAQ, freestyle, glossary, how-to guide, about us, pillar content, listicle, and comparison), set tone, length, target keywords, and special instructions, then receive a fully formatted article with SEO metadata and a featured image — ready to publish to WordPress as a post or page, export in multiple formats for static sites, or save as a content brief for a human writer. The engine draws on each client's business context (services offered, location, and brand voice) and can be enriched with optional key points, differentiators, and source material so the output reflects the client.

  GA
• Content Inventory

  Content Inventory gives teams a complete view of what already exists on a site before creating new content. For WordPress sites, it pulls pages, posts, media, and portfolio items via the REST API with full metadata (titles, URLs, post types, dates, categories). For non-WordPress sites, it discovers URLs through sitemap.xml fetching and HTML scraping — completely platform-agnostic. The inventory feeds into duplicate detection (flagging similar existing content when writing), powers the AI Content Recommendations engine with awareness of what's already published, and surfaces underserved WordPress categories with fewer than 3 posts. A search and filter UI with keyboard navigation lets teams quickly locate specific pages.

  GA
• Content Output

  Content Output is the deliverable bundle automatically generated with every article. When a post is created, the system produces: the full article HTML with proper heading hierarchy, a featured image (Gemini Nano Banana, with SEO-optimized alt text and filename), a complete SEO metadata package (meta title, meta description, keywords, canonical URL, social card tags), and permalink slug. The bundle now also includes inline JSON-LD schema markup appended to the article HTML (type-aware: BlogPosting, FAQPage (extracts Q&A pairs), HowTo (extracts steps), Service (service/city pages), DefinedTermSet (glossary), AboutPage), generated from real business profile data via utils/schema_http://generator.py. Image-text coherence: when no custom image description is provided, the image prompt is enriched with the article’s first paragraph + H2 headings so the featured image matches the article’s content. Content differentiation: the last 25 published articles for the same site are fed into the generation prompt to reduce topical repetition. Word count tracking: actual word count is stored in actual_word_count and a word-count notice may display on Post Details when outside the target range. All outputs are stored in the scheduled_posts database table and displayed on the Post Details page. For WordPress sites, the featured image is uploaded to the media library and set as the post's featured image.

  GA
• Content Authority Engine

  The Content Authority Engine is a layered prompt architecture that makes every piece of content sound human and authoritative — not like AI slop. It operates through three files: base_instructions.txt (257 lines of universal rules), type_modifiers.json (11 content-type overlays with 15-60 directives each), and prompt_http://manager.py (runtime assembly with dynamic placeholders). The system enforces: a Humanization Protocol with a zero-tolerance banned word list (30+ AI-isms like "comprehensive," "delve," "leverage," "seamlessly" are blocked), E-E-A-T Integration that dynamically injects business owner names, tenure years, and industry expertise into mandatory "Insider Tip" sections, Structural Depth Quotas that mandate H2/H3 hierarchy and section-by-section word targets for longer content, Decision Intelligence that forces clear recommendations instead of "it depends" hedging, and Conversion Architecture that embeds action-oriented narrative structure while banning generic CTAs. Anti-fabrication rules prevent invented statistics, case studies, or citations. As of June 2026, the Content Authority Shield is now visible in-product on the Post Detail page. Every completed article displays a 6-point quality audit: AI Language Filter (30+ banned phrases), CTA Prevention, Insider Expertise detection, Structural Depth validation, Active Voice enforcement, and Narrative Ending verification. Score is shown as a badge in the Quick Info sidebar and as an expandable card with per-check pass/fail detail.

  GA
• Article Creation Form

  The central interface where users configure all settings for a new article. Houses tone, length, language, permalink, keywords, scheduling, categories, special instructions, and image options. This is the cockpit — every decision about how an article gets written and published happens here. New (2026-06-24): Dotty’s Recommendations panel redesigned with a GSC-powered analysis dashboard: - GSC Hero Section: categorized Search Console queries — Quick Wins (pos 4–10), Page 2 Opportunities (pos 11–20), Content Gaps (high impressions / low clicks), Competitive (pos 20+). Each query shows position, impressions, clicks, CTR%, and actionable advice. - Keywords to Target: navigational/transactional queries are classified by intent and shown as keyword chips (not topics). Users can click chips or “Add all as target keywords” to populate Target Keywords; uses toast notifications instead of scrolling. - Source-tagged topic cards: “Targets: [query]” (purple) when mapped to a GSC query; “Search-driven” (green) when derived from GSC-related analysis. - Include branded toggle: ON by default; when off, filters out queries containing the business name and re-fetches results. - Content hierarchy + technical issues sections preserved.

  GA
• Post Details

  Post Details brings everything about a completed article into one place. It displays the full article preview, featured image, SEO metadata with copy buttons, and a quality and distribution toolkit that adapts based on site type and post status. WordPress sites see publishing controls, a live link to the page, and one-click submission to search engines for indexing. Static and non-WordPress sites see the Multi-Format Export panel (Markdown, HTML, JSON, MDX, YAML downloads). All sites see: a content quality check, a Quick Info metadata panel, Email Article Blast, a Social Content Kit for ready-to-post social content, and featured image download. The page guides you to the right next action based on context.

  GA
• Content Hub

  Client Profile parent dropdown for end-to-end blog/content management. The Content Hub is a four-tab system that covers the lifecycle of posts from draft → scheduled → published, plus Content Freshness analysis, with a shared post detail page that supports export, SEO metadata, CMS publishing, and search engine submission workflows. Child tabs included: - Drafts: write and manage in-progress articles, then continue editing in the post detail page. - Scheduled Posts: timeline view of upcoming posts with duplicate-risk indicators and basic schedule controls (edit/cancel stubs). - Published Posts: searchable, filterable library of published content with status badges and actions (details, export formats, view live, GSC stats, copy URL). - Content Freshness: passage-level AI staleness scan that flags outdated content; available on live WordPress/hosted sites only, hidden for static + project sites. Shared post detail page (/post/<id>): - Status + generation progress (published/draft/processing/failed). - Full rendered content preview + SEO metadata (multi-source fallback from Yoast/AIOSEO/AlmaSEO). - Inline quality score card (Google Cloud NLP) when Content AI is connected. - Export options (8 formats) and live URL actions when available. Integrations and delivery modes: - WordPress (full integration): publish via REST API, upload featured images, read/write Yoast + AIOSEO metadata, store wp_post_id and wp_url. - Static sites (export-based): no CMS connection required; use exports for Hugo/Jekyll/Next.js, etc. - Search engine submission: submit URLs to Google Search Console and Bing IndexNow.

  GA
• Content Extraction

  Extracts all content from a connected WordPress site via the REST API — posts, pages, media library, categories, tags, authors, navigation menus, and SEO metadata — into a browsable dashboard view and downloadable JSON file. Designed for site migrations, redesigns, and content auditing. Includes page-builder markup cleanup (Elementor/Divi/WPBakery), SEO metadata fallback via HTML scraping when plugins don't expose data to the API, and an automated extraction report explaining what was and wasn't captured. Now includes site settings (title, tagline, logo, favicon), custom post type detection and extraction with ACF/custom field support, widget and sidebar content, homepage identity scraping (phone, email, social links, copyright), image role categorization (logo, hero, headshot, background, etc.), and a content map that pre-organizes all content into homepage-ready sections (About, Services, Team, Testimonials, Contact, FAQ, Blog, Portfolio, Pricing, Careers). Data quality layer ensures all text is entity-decoded, social links are validated against the site domain, duplicate media variants are grouped with primary flags, and system author accounts are separated from real people. The Site Brief now resolves person photos even when WordPress featured_media is empty by matching image filenames to person names in page content. Now supports static/non-WordPress sites via HTML crawling. Discovers pages through sitemap.xml parsing or homepage link spidering, scrapes each page for content, metadata, images, links, embeds, and http://schema.org data, then runs the same intelligence layers (page-type classifier, image role tagging, embeds inventory, content map, site brief). Works for any website with a public URL. Static site extraction depends on the site having a parseable HTML structure. Sites with heavy JavaScript rendering (SPAs, React/Angular without SSR) may return minimal content. Rate-limited to 200 pages at 0.3s intervals to avoid overloading the target server.

  GA

The controls inside these capabilities

Finer-grained controls that live inside the capability hubs, grouped by where they sit.

Workflows these capabilities run inside

This area feeds the operator cycle — these workflows draw on its capabilities.