> For the complete documentation index, see [llms.txt](https://help.nadifor.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.nadifor.ai/ai-in-wayfor/ai-in-wayfor/knowledge-base/statuses-and-troubleshooting.md). # AI Readiness statuses & troubleshooting When you upload a document to the Knowledge Base, the platform processes it in the background so the Wayfor AI Agent can read it. The **AI Readiness** column tells you where each document is in that pipeline. {% hint style="info" %} **Who is this for?** Administrators {% endhint %} ## The statuses ![Knowledge Base Drive listing two PDFs side by side in the AI Readiness column — NIST.AI.100-1.pdf with an orange dot and "AI is learning…", EN\_US-AI-Principles.pdf with a green dot and "Ready for AI" — plus a blue "AI is learning 1 document…" banner above the table](https://1540991871-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F1lKGIWJWcNwQu3kxzBSH%2Fuploads%2Fgit-blob-2e2cb44d46346694e2c9c95d2211bfb0d2798d0d%2Fkb-drive-list-readiness.png?alt=media) | Status | Meaning | What you can do | | ------------------------------ | -------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- | | **AI is learning…** *(orange)* | The document is being processed — text extracted, indexed, and made searchable. Most files finish in under a minute. | Wait. The page also shows a banner near the top while at least one document is in this state. | | **Ready for AI** *(green)* | Processing is done. The Agent can now use this document in answers and course generation. | Nothing — it's good to go. | | **Failed** | Processing didn't complete. Usually a malformed file, an unsupported format, or a transient backend issue. | Re-upload the file, or check the document content (see below). | ## Documents stuck in "AI is learning" If processing takes longer than about 10 minutes, you'll see a warning banner at the top of the drive: ![Knowledge Base Drive with a yellow warning banner across the top reading "2 documents have been waiting over 10 minutes." and a "Retry processing →" link on the right; the AI Readiness column for both rows below is empty](https://1540991871-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F1lKGIWJWcNwQu3kxzBSH%2Fuploads%2Fgit-blob-480cec37642b6b013482364f46f9c74e9b485f67%2Fkb-stuck-banner.png?alt=media) Click **Retry processing →** in the banner to ask the platform to try again. This is usually enough — most stuck documents go through on a second attempt. If the same files get stuck again after a retry, the cause is usually the file itself rather than the platform: * **Scanned PDFs with no text layer.** The AI extracts text; if there's no text, there's nothing to learn from. Re-export the PDF with a real text layer (or run OCR on it before uploading). * **Encrypted or password-protected PDFs.** Remove protection and re-upload. * **Very large or very long media files.** Splitting a long video or audio file into shorter pieces helps. * **Corrupt files.** Try downloading the original again from its source. ## Documents that don't appear in answers If a document is **Ready for AI** but the Agent doesn't seem to use it: * **Be more specific in your question.** *"What does our handbook say about parental leave?"* steers the Agent to your KB; *"Tell me about parental leave"* may pull general knowledge. * **Confirm the content is actually in the document.** The Agent can only use what is there — check the file in **Preview** to be sure. * **Check for stale duplicates.** If you uploaded both the old and new policy, the Agent may quote either. Delete the outdated one. * **Try regenerating instead of refining.** For course generation, picking **From Knowledge Base** explicitly grounds the output in your files. From the chat, asking *"based on our Knowledge Base, …"* nudges the Agent in the same direction. ## When something looks wrong * **An old version is being quoted.** Delete the old file from the drive — the Agent stops using it almost immediately. * **A confidential document made it into the Knowledge Base by mistake.** Delete it from the kebab menu. The next answer will not include it. (For ongoing access control, the rule of thumb is: anything you put in the Knowledge Base is fair game for any admin asking the Agent.) * **A file is missing.** Check whether it was deleted — there is no trash/restore for the Knowledge Base. Re-upload from the original source. ## Related * [Upload and organise documents](/ai-in-wayfor/ai-in-wayfor/knowledge-base/upload-and-organize.md) * [Wayfor AI Agent — using the Knowledge Base](/ai-in-wayfor/ai-in-wayfor/wayfor-ai-agent/using-the-knowledge-base.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.nadifor.ai/ai-in-wayfor/ai-in-wayfor/knowledge-base/statuses-and-troubleshooting.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.