.. include:: vars.rst

.. feature_highlights:

Features
========

Files: Share Documents and Images with Connext Chatbot
-------------------------------------------------------------

The **Files** feature lets you upload and share documents directly in your chatbot conversation. This option makes it easier to tackle complex tasks like reviewing log files, analyzing design documents, or even writing code based on a photo of a whiteboard sketch.
To upload, use the paperclip icon or simply drag and drop to generate more relevant, context-aware responses.

.. figure:: _static/attachment_screenshot.png
    :alt: Screenshot showing file upload in the chatbot
    :scale: 50%
    :align: center

    Drag and drop files or use the paperclip icon to upload files directly in the chat

Supported File Types
^^^^^^^^^^^^^^^^^^^^
* **Text & Data Files** ``.txt, .idl, .json, .xml, .csv``
* **Documents** ``.pdf, .docx, .xlsx, .xls``
* **Images** ``.jpg, .jpeg, .png, .gif, .webp``

File uploads are limited to 5 per day, and the maximum file size is 20 MB. Note that images in PDF files are not yet parsed, but will be in a future release.

Memories: Create Smarter, More Personalized Conversations
---------------------------------------------------------

**Memories** let you store and manage personal preferences to streamline your experience across the entire Connext AI platform—including the chatbot, System Designer, and Copilot.

You can save preferences such as:

* Your preferred **RTI product**
* Your target **release version**
* Your go-to **programming language**

These saved preferences help |ProductName| tailor its responses to your specific development context, reducing repetitive setup and making your interactions faster and more relevant.


Creating a Memory
^^^^^^^^^^^^^^^^^

Creating a Connext AI memory is as simple as typing your preference in the chatbot; just tell it what to remember. For example:

.. figure:: _static/new_memory_screenshot.png
    :alt: Screenshot showing memory creation in the chatbot
    :align: center

    Example: Creating a memory

Managing Memories
^^^^^^^^^^^^^^^^^

After creating some memories, you can manage them in the Preferences dialog. Click your profile button in the bottom left corner of the sidebar and select **Preferences**.

.. figure:: _static/memory_button.png
    :alt: Screenshot showing the memory button in the chatbot
    :align: center
    :scale: 100%

    Example: Accessing your preferences

In the Preferences dialog, you can view, edit, or delete stored memories.

.. figure:: _static/memory_management.png
    :alt: Screenshot showing memory management UI
    :align: center

    Example: Managing your memories

#. **Product** is the RTI Connext product you want answers about, which can be Connext Professional, Connext Micro, Connext Drive, Connext TSS, or Connext Cert.
#. **Version** is the release number for your selected product (for example, 7.3.0, 6.1.2, etc).
#. **Language** is your preferred programming language. This option is especially useful for generating snippets of code.

Connext Copilot Features
-----------------------------------

To see everything Connext Copilot can do—like context-aware code suggestions, AI-powered insights, and system design—see the overview in the `Visual Studio Marketplace <https://marketplace.visualstudio.com/items?itemName=RTI.connext-vc-copilot>`_. This page includes a full list of features, screenshots, and setup instructions to help you get started quickly.

Chat with System Designer Features
----------------------------------

To explore System Designer's AI features, see RTI's documentation `here <https://community.rti.com/static/documentation/connext-dds/current/doc/manuals/connext_dds_professional/tools/system_designer/chat_sd.html>`_.

Connext MCP Server
---------------------
The Connext MCP server allows you to integrate |ProductName| into any MCP-compatible client. Your agent can send questions directly to Connext AI and use the responses as context for its responses. For example, an agent could ask a question about how to configure a specific QoS policy in Connext, and then use the response to inform its next steps in configuring a system design. We are actively working on adding more tools and capabilities to the MCP server, so stay tuned for updates!

To get started, refer to the :doc:`Getting Started <getting_started>` page.