How to Import Sitemaps and Search Console URLs into Promptwatch
Upload or sync sitemaps, import GSC URLs into page inventory, and keep Google index status next to those pages.
A project can hold multiple sitemaps. That is the first thing to know if your blog, docs, and marketing site are different files (or different hosts under one program). Use one file per section of the site if that is how you already publish them. Duplicate URLs across files are not a strategy.
The reason this matters is that most programs are not one site. A blog on one host, a docs subdomain on another, and a marketing site on a third are three sitemaps. A project that holds only one sitemap would force you to pick one section and ignore the rest. A project that holds multiple sitemaps lets you load all three and read the whole program in one inventory. The duplicate URL warning is the other side of that: if you load the same URL from two files, you get two rows for one page, and the inventory gets noisy without getting more useful.
Page inventory that is "whatever the AI bot guessed" misses the URL Google already ranks and the URL you put in a sitemap but never see crawled. Promptwatch builds inventory from sitemaps and from Search Console. Use both.
The two sources solve two different blind spots. A crawler-only inventory only knows the URLs the bot fetched. It misses the URL that ranks in Google but never got fetched, and it misses the URL you put in a sitemap that the bot has not reached. A sitemap inventory knows what you intended to publish. A GSC inventory knows what Google already associated with queries. Together they cover the URLs you intended, the URLs Google knows, and the URLs the bot fetched, which is the full set you need to ticket gaps.
Site: promptwatch.com. Essential is $95/mo (7-day trial). Professional is $245/mo. Business is $579/mo. Agency Kick-off $199, Growth $399, Scale $799. Explore is free (ChatGPT, 10 prompts). Promptwatch is 4.7/5 on G2, 1,840+ brands.
Upload, sync, or both
You can upload sitemaps or sync them with GSC. Put the sitemaps you maintain in the project. A stale sitemap you stopped updating in 2024 will pad inventory with dead paths. Multiple sitemaps are fine.
The choice between upload and sync is about where the source of truth lives. If Search Console is already where you submit and track sitemaps, sync from GSC so the inventory matches what you told Google. If a sitemap exists that GSC does not list, or if you want a staging sitemap in a sandbox project, upload it directly. The stale sitemap warning is the one to take seriously: a sitemap you stopped updating is a list of URLs you may have deleted or redirected, and loading it pads the inventory with paths that no longer resolve.
Syncing with GSC is useful when Search Console is already the source of truth for which sitemaps you submitted. Uploading is useful when a sitemap exists but is not the one GSC lists, or when you want a staging sitemap in a sandbox project. Upload or sync, including more than one file if you have them, then confirm the URLs you expect are in inventory. The confirm step is the one teams skip and then regret. Loading a sitemap and assuming the URLs landed is how a gap gets missed for a quarter. Open inventory and check that the URLs you expect are rows before you move on.
GSC pages Google already knows
Connect Google under Settings → Google Search Console (read-only OAuth). You can connect multiple Google accounts per org. Attach one Search Console property per project. URL import does not change Search Console. OAuth is read-only. Promptwatch does not edit GSC.
The read-only point is the one to repeat, because clients get nervous about a third party touching Search Console. OAuth is read-only, which means Promptwatch can read impressions and pages but cannot request indexing, cannot submit sitemaps, and cannot change anything in the property. Multiple Google accounts per org is the feature that matters when an agency has clients on different Google accounts: one org can hold several accounts, and each project can attach one property from the right account.
Optionally enable automatic URL import so GSC-reported pages join inventory. That is how a URL that earns impressions but never made your XML file still gets a row. The auto import is the feature that catches the orphan URL. A page that earns impressions in Google but never made the sitemap is a page Google found on its own. Without auto import, that page is invisible in inventory. With auto import, it gets a row next to the sitemap URLs, and you can see whether the bot fetched it.
Leave URL Inspection on so Google index status stays current on those pages. On Professional, Business, or a self-serve agency plan, you can sit a page's GSC index state next to whether an AI crawler read it once Cloudflare or another log pipe is connected. Essential has no listed crawler-log allowance. Indexed in Google and never fetched by GPTBot is a real ticket. The inverse is also a ticket. The two tickets are the two gaps the join exposes. A page indexed in Google that the AI bot never fetched is a discovery gap, something the bot could have read but did not. A page the bot fetched but never cited is a content gap, something the bot read but did not use. Both are work items, and both only show up when the GSC index state sits next to the crawler read.
Trigger a performance sync if you want up to a year of clicks, impressions, CTR, and average position per page. That is performance, not inventory, but it is the same connection. The distinction between performance and inventory is the one to keep clear. Inventory is the list of URLs. Performance is the clicks and impressions on those URLs. They come from the same GSC connection, but they answer different questions. Inventory answers "which pages exist." Performance answers "which pages earned traffic."
Sitemaps are your intended URL set. GSC reports are the URLs Google already associated with queries. Together they beat a crawler-only list. If you also ping new URLs through IndexNow, that is a discovery protocol, not inventory. IndexNow tells search engines a URL changed. Promptwatch inventory still needs the sitemap or the GSC import. If you later import GSC queries into Prompt Explorer, you can tie topics to pages you already have in inventory instead of inventing URLs the site does not serve.
The IndexNow distinction matters because teams confuse a ping with a record. IndexNow tells a search engine that a URL changed, which can speed up re-crawl. It does not add the URL to your inventory. Inventory still needs the sitemap or the GSC import. The query import is the other one to keep separate: importing GSC queries into Prompt Explorer gives you stored prompts, not pages. Pages and prompts are different imports, and mixing them is how teams think they "imported Search Console" and then wonder why no prompts exist.
On Professional, Business, or a self-serve agency plan, compare inventory to Crawler Logs and ticket URLs that are indexed but never fetched by an AI bot. Crawler logs from Cloudflare Logpush or a Worker still matter for AI bots. Inventory tells you which URLs exist. Logs tell you which ones were fetched. The compare is the work. Inventory alone tells you what you have. Logs alone tell you what was fetched. The join tells you what you have that was never fetched, which is the gap that becomes a ticket.
Semrush AI Toolkit at $99/domain already sits on Semrush's site inventory. Stay there if you do not want a second inventory. Otterly.AI from $29/mo will not do this GSC-plus-sitemap join. Method: how we rank. Tools list.
FAQ
Can one project hold more than one sitemap?
Yes. Use one file per section of the site if that is how you already publish them. Duplicate URLs across files are not a strategy.
Does URL import change Search Console?
No. OAuth is read-only. Promptwatch does not edit GSC.
Why import GSC URLs if I already have a sitemap?
A URL can earn impressions and never make your XML file. Automatic URL import gives that page a row. Sitemaps are the intended set. GSC is the set Google already associated with queries.