RAG Chat

The chat that knows what it is talking about

Ask questions and receive answers with cited sources from all of your local knowledge: emails, documents, notes, Kanban cards, and meeting minutes.

Streaming answers Cited sources Search scope can be restricted Conversation history

The heart of Alcilla

The chat searches not just one mailbox or folder but the entire local index—and says so openly when the requested information is not present in the documents.

Streaming answers with sources

Answers appear word by word, with source chips underneath. One click opens the original context.

Streaming answers with sources

Restrict the scope precisely

The scope picker limits the search to specific folders or sources.

Restrict the scope precisely

Check Sources

Sources can be accessed directly from the chat.

Check Sources

Open Sources Directly

Click to jump straight to the relevant passage in the original source.

Open Sources Directly

Selectable model & profile

Multiple language models or endpoint profiles can be configured and selected as required.

Selectable model & profile

A genuine conversation

Follow-up questions refer to the previous conversation, and long histories are summarized automatically.

A genuine conversation

Ratings & cache

Answers can be rated with a thumbs-up or thumbs-down, while very similar questions can be answered immediately from the cache.

Ratings & cache

Why evidence-based answers make a difference

A chatbot that merely sounds plausible is not enough for day-to-day work. Alcilla consistently prioritizes traceable sources over convenient guesswork.

Honest answers

The chat stays within what it can find in your documents and states clearly when no relevant information is available.

‘Without search’ as a deliberate exception

For general questions unrelated to your documents, document search can be disabled intentionally. You remain in control.

Everything stays local

Search, index, and model run in your environment. In the local operating model, content is processed exclusively through systems configured by the customer.

From a question to an evidence-based answer

  1. 1

    Ask a question

    Ask a question

    Use the input field or start directly from email or Kanban.

  2. 2

    Search the index

    Search the index

    Semantic search across all approved sources.

  3. 3

    Generate the answer

    Generate the answer

    The answer is grounded in the passages found.

  4. 4

    Review sources

    Review sources

    Each chip opens the original context for verification.

One index for all Alcilla tools

RAG Chat is not a separate tool. It searches the same content already known to email, Kanban, notes, and the AI Editor.

Email Kanban Notes Documents Minutes

Experience RAG Chat live

Ask your first question of your own company knowledge.