The sidebar is your main navigation hub — present on the left side of every screen.

➕ Create The + icon lets you quickly spin up something new — a fresh conversation thread, a new agent, or another resource. The fastest way to get started on something from anywhere in elvex.

Tip: Start new threads often.

💬 Threads Your full conversation history. Every chat you've had with any agent in elvex is stored here and searchable at any time. Use it to pick up where you left off, reference past work, or find a thread to share. Threads is also where you'll find any scheduled agent runs you've set up.

Tip: When you start a new thread make sure you assess the chat box.

📦 Resources Your library of everything available to you in elvex:

Tip: Use a the sidebar to find and click on a resource to add it to the context strip.
​

🗂️ My Space Your personal workspace in elvex. It holds your personal context (background about you — your role, preferences, and priorities — that flows into every conversation automatically), any agents, datasources, and threads you've pinned for quick access, and any sub-spaces you've created for specific projects or initiatives.

Tip: Click Configuration in order to add and edit your personal context.

🔔 Notifications The bell icon surfaces updates relevant to you — like when a shared resource is updated or a scheduled agent run completes.

🔍 Universal Search The magnifying glass lets you search across everything in elvex — threads, agents, datasources, artifacts, and more — so you can find what you need without navigating manually.

🛡️ Admin (admins and owners only) The shield icon provides access to company-wide management tools: team member management, integration authorization, the audit log, and company-wide settings.

The chat box is where every conversation begins — and it's more than just a place to type.

At the top of the chat box, the context strip shows what foundation your conversation will start from. This might be your personal context, a Space's context, or a specific agent. What you select here shapes the entire thread — it determines what background knowledge elvex is drawing on from the very first message. Once you send, the context strip moves up to the thread header so you can always see what context your conversation is grounded in.

To the left of the input field, the plus icon lets you attach resources to individual messages — agents, datasources, spaces, threads, or files. Unlike the context strip (which applies to the whole conversation), attachments are message-level: you're bringing something in for a specific question or task, right now.

Together, the context strip and attachments give you precise control over what elvex knows and what it's working with at any point in the conversation.

🧠 Thinking When enabled, elvex reasons through your question more deliberately before responding — working through the problem step by step rather than answering immediately. Useful for complex questions, multi-part tasks, or anything where you want a more considered answer.

✏️ Off the Record When enabled, this message won't be used to inform or update elvex's understanding of your context. Useful for one-off or sensitive questions where you don't want the exchange to influence how elvex thinks about you going forward.

Together, the context strip, attachments, and toggles give you precise control over what elvex knows and how it responds at any point in the conversation.

Once you kick off a conversation, icons appear on the right side of the thread. This is your in-conversation hub — giving you quick access to two things relevant to that specific thread.

⚛️ Integrations This panel shows the integrations available during this conversation — connections to external tools and services like Slack, Gmail, Salesforce and more. When an agent has access to an integration, it can take actions in those tools on your behalf: sending an email, creating a record, posting to a channel. The integrations panel lets you see what's connected and, in some cases, manage authorization directly.

Artifacts This panel shows any files or outputs that have been generated during the conversation. When an agent produces a document, spreadsheet, presentation, or other file, it surfaces here as an artifact — ready to preview, download, or share. Artifacts persist beyond the conversation, so anything generated in this thread is saved and accessible even after you leave. You can also return to artifacts using Universal Search.