Edit Flow stores custom statuses in the wp_posts table's post_status column (values like pitch, assigned, draft, pending-review) and keeps editorial metadata in wp_postmeta under keys prefixed _ef_editorial_meta_. The story budget pulls all of that together into a single overview that editors normally click through one section at a time.

SleekAI maps those exact tables and meta keys into the chatbot's system message. Ask which articles are stuck in pending-review, which writer owns the homepage feature for Friday, or how many pieces are sitting at assigned with no first draft, and the bot pulls the answer from live data. Editorial comments stored on the post are also readable, so it can summarize the last note an editor left on a draft.

Generic chatbots can't see any of this because Edit Flow's statuses are custom, not WordPress core. They also can't read the _ef_editorial_meta_ postmeta the team relies on for deadlines, word count targets, and assignment notes. SleekAI's display conditions can scope the bot to the admin area only, so freelancers without dashboard access never see it.

Edit Flow has been a quiet workhorse for editorial teams since 2009. It does one thing very well: keeps a small group of editors and writers honest about what's pitched, what's drafted, and what's actually ready to publish. The catch is that all of that state lives in custom post statuses and a quiet set of postmeta keys that no generic chatbot will ever recognize.

A bot that says "I don't see any drafts" because it only knows the publish/draft/private values WordPress ships with is worse than no bot at all, because editors will stop trusting it on day two. SleekAI reads the real Edit Flow tables directly, including the custom statuses you configured for your team. Monday status meetings get shorter when one editor can ask "what's in pending-review with photos still open" and get a clean list instead of clicking through the story budget.

Freelance writers don't see the bot because display conditions scope it to admin users with the editor role or higher. Token spend stays low because most queries are short factual reads against your own database, and you bring your own API key so there's no per-seat markup. The bot also doesn't try to write to the workflow unless you explicitly enable agent mode.

For a small newsroom, that read-only default is what makes it safe to leave running all day.