The ways in

From a URL

Point the desks at a page you already have - your own ageing post or a page you want to beat - and get a new source-backed article out of it. Works for one page or a whole sitemap.

Point the desks at a web page and they write a new article from it. Use it for three jobs:

Refresh - an older post of yours that has gone stale, brought up to date.
Rewrite - the same subject, written again in your voice and to your standard.
Outperform - a page that ranks where you want to rank. The desks write an article meant to beat it.

The page is the starting point, not a template. The desks read it, research the subject as they would for any researched article, and write a new piece. How much of the finished article matches any source word for word is one of the article's recorded checks, so you can see it.

On this page

One URL, or a list

Type one address for one article. Press Enter for a second line and it becomes a run, one article per address - the same gesture that turns one topic into an Assignment.

https://example.com/guide-to-composting
https://example.com/best-garden-tools
https://example.com/how-to-prune-roses

Nothing is sent unread

One URL or twenty-five, every page is fetched and shown to you before anything is spent: the headline found, the site, the word count and the date. A page that could not be read is named and left out of the run, never written against.

For a list this is a checklist. Every page arrives ticked; untick the ones you do not want. One selection decides both what gets read and what gets sent, and it survives the read - so pruning a long list is work you do once.

The read step catches what an address alone cannot:

a dead link
a paywall, or a page behind a login - these give up little or no text
a page drawn entirely in the browser, which a reader can see but a fetch cannot
a section index or author page - a list of headlines, not an article. A sitemap will happily include these. When one yields text it is shown with what it yielded, and the decision is yours.

Upload a list, or your sitemap

Press Upload to read .txt, .csv, .xlsx, .docx, .pdf or .json files. Only lines that hold a web address become rows, so a document with no addresses in it adds nothing and says so. An export with a title and an address on each line is read whole, wherever the address sits.

Upload also reads a sitemap. Give it yoursite.com/sitemap.xml and it lists what your site publishes. You can narrow it:

to a path, such as /blog/
to pages not updated in the last 3, 6 or 12 months
to the first N pages

It reports three numbers - how many pages the sitemap lists, how many survived your filters, and how many are being taken - because a filter that quietly ate the list looks exactly like one that worked. A sitemap that points to other sitemaps is followed one level down.

This is the quickest way to find the posts on your own site that are due a refresh.

What it costs

Every page is a full article, written by all ten editorial desks. Ten addresses is ten articles, not one cheap batch. A run is up to 25 pages, and the same address twice is written once.

From a URL or Dossier?

From a URL - start from this page, and research the subject around it.
Dossier - write from the pages you attach and nothing else.