> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mithunai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Ask questions grounded in your knowledge in Chat

> Chat's Knowledge tool lists, searches and asks your assistants over MCP, returns cited answers or says it does not know, and cannot change anything.

MITHUNAI Chat answers from your own knowledge through the **Knowledge** tool: the same three
read-only tools the [MCP server](https://docs.mithunai.com/channels/mcp.md) offers, connected to Chat for your organisation.
Switch it on in a thread and the model can list your knowledge, search it for passages, and ask
one of your assistants -- which answers with citations, or says your content does not cover the
question.

## Switch it on in a thread

1. In a thread, open **MCP Servers** under the message box.
2. Select **Knowledge**.
3. Ask your question. Saying "using our knowledge" or "from our documentation" makes the model reach
   for the tool rather than answer from its own general knowledge.

**Knowledge** is the only server in that menu, and you cannot add another: members cannot connect
their own MCP servers to Chat.

## What the Knowledge tool does

| Tool | What it does | Counts against monthly answers |
| - | - | - |
| `list_knowledge` | Lists the assistants you can ask and the knowledge bases you can search | No |
| `search_knowledge` | Returns the passages that best match a query, each with its source. Generates no answer | No |
| `ask_assistant` | Asks one of your assistants. Returns a grounded answer with citations, or an abstention when your content does not cover it | **Yes**, one answer per call |

The model chooses which tool to call, and may call several in one reply -- listing first, then
asking. Each call appears in the reply as a step you can expand to see exactly what the tool
returned. Each `ask_assistant` call is also saved as a one-question conversation in the console's
[Conversations](https://docs.mithunai.com/analyze/conversations.md), so it shows up in your analytics like any other answer.

## How a grounded reply differs from an ordinary one

Chat is a general chat workspace, so not every reply comes from your knowledge.

| The reply came from | What it is | Citations |
| - | - | - |
| `ask_assistant` | Your assistant's answer, written only from your content, by the same [answer engine](https://docs.mithunai.com/concepts/how-answers-work.md) as the widget and the API | Yes |
| `search_knowledge` | The model's own writing, from passages the search returned | The passages' sources |
| No Knowledge tool call | The model's own answer from its general training | No |

**When an answer must be correct to your documentation, check that the reply used the Knowledge
tool** -- the expandable step shows it -- and open the sources. For answers where grounding is
guaranteed on every turn, use the console's **Playground**, the [website widget](https://docs.mithunai.com/channels/widget.md) or
the [HTTP API](https://docs.mithunai.com/channels/api.md).

When your content does not cover a question, `ask_assistant` returns an abstention rather than a
guess. A reply that says your knowledge does not cover it is the tool working as designed, and a gap
worth filling; [Coverage gaps](https://docs.mithunai.com/analyze/coverage-gaps.md) lists them.

## What it will not do

* **Change anything.** It reads with the access of a **Member**: it cannot add, edit or re-sync a
  source, change an assistant, or touch a setting.
* **Reach another organisation's knowledge.** Your Chat serves one organisation, and the key it uses
  belongs to that organisation; MITHUNAI derives the organisation from the key, never from anything
  Chat sends.
* **Read as you personally.** Everyone in your Chat reads the same knowledge, through one
  organisation key. Your own role in the console does not widen or narrow it.
* **Search the web.** Web search is switched off unless your deployment has set it up. See
  [What is switched off](https://docs.mithunai.com/chat/administration.md#what-is-switched-off-and-why).

## Files you attach

A file you attach to a message -- text, Markdown, CSV, PDF or an image -- joins that turn's context so
the model can read it. It is **not** added to your knowledge and is not searchable by the Knowledge
tool. To make content answerable everywhere, [upload it as a source](https://docs.mithunai.com/knowledge/file-uploads.md) in the
console.

## Limits

* **`ask_assistant` waits for the whole answer.** It does not stream, so a long answer appears all at
  once. A call that takes longer than 60 seconds is abandoned.
* **Answers count.** Each `ask_assistant` call uses one of your organisation's
  [monthly answers](https://docs.mithunai.com/analyze/usage-and-limits.md); at the limit it is refused, and the reply says so.
* **The chat model writes the final reply.** It can summarise or reword what the tool returned. The
  expandable step is the record of what your knowledge actually said.

## Related

* [MCP server](https://docs.mithunai.com/channels/mcp.md) -- the same tools, from your own AI client.
* [Configure an assistant](https://docs.mithunai.com/assistants/configure.md) -- what `ask_assistant` asks.
* [How answers work](https://docs.mithunai.com/concepts/how-answers-work.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.