SemAgentSources
Admin Panel → Sources shows where your knowledge base content came from. Knowledge Base shows documents and Crawl shows the process; only this screen answers "which address produced how many documents?".
blog.example.com is configured as a source but has never put a document into the knowledge base.What it shows
Each row is an origin: a sitemap, a single URL, a store page, the files you uploaded or the text you typed in. The row tells you how many documents that origin produced, when its latest content arrived, and its crawl schedule if it has one.
| Kind | What it means |
|---|---|
| Sitemap | Addresses crawled through a sitemap.xml. |
| Single URL | One page you added by hand. |
| Store page | Policy / FAQ / about pages pulled automatically from a connected e-commerce store. |
| Uploaded file | PDF, Word, Excel, CSV, text and similar uploads. |
| Typed text | Content written straight into the Knowledge Base. |
The most important thing this screen can tell you
"Configured but produced nothing." If an origin is set up yet no document ever reached the knowledge base, the row says so plainly. The crawl may never have run, the address may be unreachable, or the page may be empty. No other screen shows this — in the Knowledge Base you see the documents that exist, never the ones that do not.
Nothing is written here
Sources is a derived view: the numbers are computed from knowledge base records and nothing is changed on this screen. Use Knowledge Base to add or remove content and Crawl for schedules; the shortcuts in each row take you there.
Not to be confused with its neighbours: Knowledge Base shows documents, Crawl shows the process, Data Sources shows external integrations. Only this screen answers "which address produced how many documents?".