The blog posts Pendium generates for your Content Pipeline are substantially different starting with this release. Each post now opens with a dense lead paragraph written for AI passage retrieval, uses real heading hierarchy throughout, and anchors major sections back to your brand. Posts include Markdown tables where they help, cite two to four external sources, and pull relevant photos inline. The JSON-LD on every article now carries entity signals, and genuine Q&A posts emit FAQPage schema — both of which improve how AI platforms read and attribute your content.
The quality review step in the workflow was also quietly underperforming. It was flagging claims and links as unverified even when the writer had direct sources for them, because the reviewer wasn't receiving that source material. The reviewer now sees the same research the writer used — RAG results, web research, scraped pages, search results, your Key Pages — so false flags on sourced claims and valid URLs drop significantly.
During onboarding, the first piece of content now runs through the full cell pipeline in the workflow view — outline, text, review, media, publish — so you can watch each stage advance live. Previously the post would appear in a separate list with no workflow context, which made the connection between content and pipeline invisible. The workflow also no longer generates a second batch of ideas after you complete the wizard; it opens with the five you selected, not ten.
A few smaller corrections worth noting: brand colors extracted from your site now pass a contrast check before being applied to agent pages and your hosted blog, so saturated hero colors no longer make body text unreadable. The Company Profile enrichment now pulls in your Visibility Scan personas as context, so the two stay meaningfully connected without merging into the same list. Adding a page in Key Pages now inserts the new row at the top, so it's visible on mobile without scrolling down.