> ## Documentation Index > Fetch the complete documentation index at: https://docs.getvector.me/llms.txt > Use this file to discover all available pages before exploring further. # Competitor content archive > Save a competitor's help center or changelog as a complete copy, read it inside Vector, and see exactly which pages changed over time. A [competitor](/competitors/add-competitors)'s help center says how their product really works, and their changelog says what they've shipped. The content archive keeps a complete copy of both inside Vector: you can read it even after a page is changed or deleted, cite it as proof, and see exactly which pages changed between visits. The archive works for the **📚 Help/Docs** and **📝 Change log** tabs of a [competitor dossier](/competitors/competitor-dossier). Other sections, such as the blog or release notes, are planned to follow. ## Archive a help center or changelog In the competitor's dossier, open **📚 Help/Docs** — or **📝 Change log** for their changelog. Click **Add Help/Docs source** (or **Add Changelog source**), and enter the address the section starts from — or confirm the one Vector suggests. Click **✦ Crawl now**. The crawl runs in the background, so you can leave the page. Pages appear in the tab as they land. Crawls need a Firecrawl key, on the [Web scrapers](/ai/web-scrapers) tab of **Settings**. A crawl captures up to 500 pages by default. For a larger portal, raise **Max pages** — up to 5,000 — before you crawl; pages beyond the limit aren't captured. The tab lists every captured page by title and path. ## Read a saved page Open any page in the list and it shows inside Vector with its text, layout and images, even if the live page is down or has since changed. Alongside it you see when it was captured, which tool captured it, a link to the original, and how far the capture can be trusted. Links to other pages of the same section open their saved copies when those were captured too. A link to anything outside the archive is marked as leaving it, and opens the live page. ## See how a page changed Each new capture of a page is kept as a new version; the earlier ones stay. In the page viewer, pick a version to see the page as it was then, and **diff vs previous** to see exactly what changed since the capture before it. ## Bring the archive up to date Click **✦ Crawl now** again. Pages that changed since the last crawl get a new version, so their history grows; new pages join the list. A page that's gone from the competitor's site stays in the archive, marked **removed** in the list, so you keep it as evidence. ## Choose how carefully changes are checked **Trust mode** sets the balance between cost and certainty: * **Lightweight — cheap (hash + plausibility)** — the default for Help/Docs. Fast, and enough for most portals. * **Full corroboration — pricier, fewer risks** — confirms a change against a second, independent source before trusting it. Switch at any time; the tab explains the trade-off before you do. Use full corroboration on the pages where a wrong alert would cost you most. ## Use their docs as feature evidence Once a section is archived, Vector looks for passages that describe how the competitor implements a feature you track. They show on that feature's dossier, on **Overview** → **Archive**, under **📚 How it works — from competitors' docs**, each with an excerpt, why it matched, and **open saved page↗**. Each passage starts as **proposed**. **Confirm** it to make it part of the feature's evidence, or **Dismiss** it. Your decisions stick: matching again later doesn't undo them. ## Where to go next Your own material, alongside theirs. Getting told when a competitor changes something.