> For the complete documentation index, see [llms.txt](https://docs.katara.ai/about/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.katara.ai/about/documents-and-collections/curated-answers.md).

# Curated answers and questions

Use **Q\&A Dashboard** to maintain reusable answers and their associated questions. This is separate from configuring a [Q\&A agent](https://docs.katara.ai/about/agents-and-answering/qna.md), uploading source documents, or editing Brain knowledge.

## Before you start

Select the intended organization and confirm you can create or update Q\&A records. Decide which scope the answer belongs to before saving: a well-written answer in the wrong scope may not be used by the workflow you test.

## Create an answer

1. Open **Q\&A Dashboard → Answers**.
2. Select **Create a new answer**.
3. Complete **Answer**, add the intended **Questions**, and review **Sources** and **Scope**.
4. Select **Create answer**.
5. Find the saved answer using the available search, scope/status filters, and sorting controls. Load more results when needed.
6. Review the answer and associated questions. Use the answer's edit control and **Update** to correct it, or **Add question** to associate another wording.

Check reactions and usage metadata as supporting feedback, not proof that every user received the intended answer. Test representative wording with the relevant agent in [Prompt Lab](https://docs.katara.ai/about/agents-and-answering/prompt-lab.md).

## Choose a question type carefully

Under **Questions**, **Create a new question** opens a form with **Type**, **Question**, **Answer**, and **Scope**. Select or create the associated answer as appropriate, then select **Create question**.

* **Cached (default)** reuses the associated answer when the question is matched.
* **Blocked** prevents the question from being answered.
* **Real-time**, when exposed, is currently equivalent to blocked. Live-API answering is not implemented here. It is not a bypass-cache or fresh-generation mode.

## Import and export spreadsheets

Open **Import and Export**. Start with **Download template**, fill it using the provided structure, then select your file through **Import data**. Review the **Questions** and **Answers** preview and validation errors before selecting **Start import**. Inspect the per-item status and logs afterward; do not assume every row succeeded because the import finished.

Use the export controls to export Q\&A data. This workflow does not export Brain Wiki history to GitHub.

## Recover from mistakes

For a missing record, check filters, scope, and additional result pages. For a failed import, correct reported rows and compare existing records before retrying to avoid unintended duplicates. Reopen edited items to verify saved content. If a question unexpectedly stops answering, inspect its type first, especially an existing **Real-time** value.
