Files, text, Q&A and website pages

Everything your AI reads that is not a knowledge base or a documentation set — uploaded files, pasted text, question-and-answer pairs and pages of your website. It all belongs to the brand, and it sits in collapsible groups on the Knowledge screen.

Where these live

Open Brand → (your brand) → Knowledge. The page is headed What Yaplet answers from, and under its summary the Sources heading introduces everything Yaplet reads. First comes the Knowledge bases & documentation group — the material that is also published for your customers to read. Below it come the groups this page is about. None of them appears on your public site: they are knowledge only, used to answer in chat, on the phone and in social posts. The marketing autopilots (Social Posting's Autopilot and Newsletter Autopilot) write from these sources only when they are ticked for them on the brand's Content sources page.

There are no tabs. Each kind of source has its own group, which opens and closes with a click and appears as soon as you have something of that kind. A group's header counts its sources and shows how many pieces are indexed, or how many need attention, and a group that needs attention opens by itself.

  • Website — your websites. Yaplet checks them every hour, so each site gets its own kind of row with its own controls; they are covered on their own page.
  • Files — documents you upload.
  • Text snippets — text you paste in. It never changes on its own.
  • Questions & answers — a question paired with the exact answer you want given.
  • API tools — endpoints the AI can call for live data, covered on their own page.

Everything here belongs to the brand, not to the AI agent. Replace or delete the agent and these sources stay exactly as they are, and websites keep being checked on their normal schedule.

What each group shows

Websites sit apart from the rest because they behave differently: Yaplet checks them every hour, they can fail, and one site is many pages. A website row therefore says how many of the site's pages are learned and when it was last checked, and has a Manage pages button; see the Website pages article below.

The other groups hold things that are simply stored as you left them. Each row shows the source's name, a line of facts under it (the kind of source, how many pieces are indexed — or Not indexed — and the date it was added) and a status badge. Once a group has more than ten rows, a search box (by name) and a status filter appear above its list, and the list is paged.

Every row has an Actions menu (⋯) that offers only what applies to that row: See what was stored, Edit (text and Q&A only) and Delete. A website row has a Manage pages button, and its menu offers Check now, Open site, Pause automatic checks or Resume automatic checks, and Delete. The Website group's own ⋯ menu (Website actions) has Read all again, which queues every website for a fresh read.

An older single web page or sitemap that was added before websites became one source each keeps its own kind of row: it says Web page or Sitemap, when it was read (Read 3 days ago) and when it is read next, and its menu offers See what was stored, Open the page, Read again now, Pages found (sitemaps only), Pause automatic re-reading or Resume automatic re-reading, and Delete. Pages found opens a list of the sitemap's pages, headed N pages were read on the last pass and M could not be; hover Failed on a page to see why.

Adding a source

Everything is added from one place: the Add knowledge button at the top of the page. It opens a window with two groups.

  • Teach Yaplet — Website, Files, Text snippet, Question & answer and API tool.
  • Help content customers can read too — Knowledge base and Documentation.

Choose one and follow the steps. A website only needs its address; the Website pages article below explains what happens next. The other windows ask for:

  • Add a text snippet — a Title (how the snippet is listed on the Knowledge page) and the Text itself; press Add snippet. Two snippets cannot share a title.
  • Add a question and answer — a Question and the exact Answer Yaplet should give when a customer asks it; press Add answer. The same question cannot be added twice.
  • Upload files — drag and drop or choose PDF, DOCX or TXT files, and give the upload a Name (how it is listed on the Knowledge page); press Upload. Yaplet reads the text in the documents and answers from it. A file with the same name cannot be added twice.

To change a text snippet or a question and answer later, choose Edit in its row's menu: the window is called Edit text snippet or Edit question and answer, and Yaplet re-reads it as soon as you press Save changes.

Deleting a source removes it and everything the AI learned from it. The confirmation says so: "Delete “the source's name”? Everything Yaplet learned from it is removed at the same time."

Status — is this source actually working?

The status badge on each row is the answer to "why isn't my AI using this?".

It can read:

  • Ready — the AI can answer from this right now.
  • Not indexed — nothing was stored from it, so the AI cannot answer from it. Usually an empty file, or a page that could not be read.
  • Reading… — we are working through it at the moment.
  • Warning — it was read, but something went wrong along the way.

Files, text and Q&A can also show Failed — the last attempt to store it did not succeed.

Websites have one of their own instead: Stopped. A site that cannot be read is tried again every hour, and only when its list of pages has been unreadable for 14 days do the hourly checks stop. Your AI keeps answering from what it already learned, the account owner is emailed after 24 failed checks, and Check now tries again.

Hover a badge to see the reason. A group that holds a source needing attention opens by itself when you open the page, so you are not left to discover it by scrolling.

Indexed — how much a source contributes

Each row says how many searchable pieces the AI stored from that source (for example 12 pieces indexed), and the group header and the summary at the top add them up. Longer sources become more pieces. Not indexed means nothing was stored, which is the same thing the Not indexed status is telling you.

Two things surprise people here. One article usually becomes several pieces, so the number is always larger than the number of documents you added. And for knowledge bases and documentation sets, only published articles are stored — a draft contributes nothing.

Product sources are not here

A product source is not listed among these sources, so the same thing cannot be deleted from two different screens. Product sources — their refresh status, their last error and how many products came out of them — live only on Brand → Products.

What language a source is searched in

A source's language decides which dictionary is used when the AI searches its text. It does not translate anything.

Knowledge bases and documentation sets each carry their own language, set on the row in the Knowledge bases & documentation group. The sources on this page — files, pasted text and Q&A — carry no language of their own and fall back to the brand's language, set at Brand → Brand settings → General. A website is different: each site has its own language choices, made in Manage pages and covered in the Website pages article.

The AI searches in every language the brand's content is written in, so a brand holding an English and a Hungarian knowledge base keeps full accuracy in both.

The indexed-item allowance

Money decides what Yaplet may learn, and nothing already taught is ever removed.
  • On a paid plan there is no limit on how much Yaplet learns, except 2,000 pages per website.
  • During the free 14 days, or with bought credit and no plan, no limit shows on screen, but one exists. If you reach it, the website row says You've reached your knowledge limit with a Write to us link.
  • If nobody can pay (no plan, no bought credit, and the free 14 days are over), nothing is learned or refreshed, Yaplet's answers are paused, and the payment window opens when you try to add something. The summary at the top says Paused until you add credit or choose a plan.
  • Some older free workspaces keep a limit of 250 searchable pieces across the whole workspace. Only they see a count (212 of 250 on the free plan) and, when it is full, Your free plan's knowledge is full. At the limit, publishing new articles and adding new sources is paused.
See Free plan limits for the full breakdown.

Who can see these groups

The Knowledge page opens for anyone holding Content sources, Knowledge base or Documentation. The groups on this page only render for team members holding Content sources — so a colleague who only manages help articles sees the Knowledge bases & documentation group and nothing else.

Website pages

Add your website once and Yaplet learns its pages and keeps up with every change. Covers choosing which pages to keep, languages, how the hourly checks work, and who pays.

API tools

Let the AI fetch live data while it answers — order status, stock, account details. Supports forms that collect what the call needs, and handles failures gracefully.

Best practices

Give sources clear names. The list is searchable by name, and a list of files called export-final-2.pdf is a list you cannot manage.
Quality over quantity. Focus on high-quality, relevant material rather than adding everything you have. Better sources make better answers.
Check the status after adding anything. A source sitting at Not indexed contributes nothing, and nothing else on the screen will tell you.

© 2026 Yaplet. All rights reserved.