> For the complete documentation index, see [llms.txt](https://docs.nimbalyst.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nimbalyst.com/session-management/agent-window-and-session-management.md). # Agent Window & Session Management Manage AI coding sessions in the Nimbalyst Agent window. Start, resume, group, rename, and search long-running Claude Code and Codex sessions. The Agent window is where you run and manage your AI sessions: long chat discussions, research, complex multi-file tasks, and AI-assisted development. Where Files mode is for editing your documents, the Agent window is for working with your agents and keeping many sessions organized.

The Agent window with the session list on the left, a transcript in the center, and the session panel on the right — The Agent window keeps the session list, conversation, and session details together.

The Agent window has three main areas: * **Left sidebar** with your sessions. Search and select sessions here. * **Central transcript** where you converse with the agent, with full conversation history, tool call visualization, and streaming content. * **Right panel**, which you can switch between edited files, review, and a chat about the session. The [session management feature page](https://nimbalyst.com/features/session-management/) shows how sessions, [workstreams](/session-management/workstreams.md), and the kanban view fit together. ### Starting Sessions Click the main **New session** button in the title bar, or press **Cmd/Ctrl+N**, to start a regular session immediately. When other launch types are available, the adjacent menu lists options such as **New Worktree**, **New Blitz**, **New Terminal**, **New Super Loop**, and **New Meta Agent**. Which options appear depends on Developer Mode, whether the project is a git repository, and the alpha features you enabled. To add a related session next to one you already have, use the **+** next to the session tab instead; see [Workstreams](/session-management/workstreams.md). ### Choose What the Right Panel Shows The right panel is a switcher. Use the dropdown in its header to change what it displays.

The agent window right panel dropdown, open on three options: Edited Files with a checkmark, Review, and Chat with Session — Switch the right panel between edited files, review, and a chat about the session.

* **Edited Files**: the files the agent created and changed in this session, with uncommitted changes and commit controls. See [View Files in Agent Mode](/session-management/view-files-in-agent-mode.md). * **Review**: the review queue for changes waiting on your approval. * **Chat with Session**: a conversation about the session itself, kept separate from the main transcript. You can also open a tracker item as a tab in this window and work the item and the session together. See [Item Detail](/task-management/item-detail.md). ### Finding and Organizing Sessions Run as many parallel sessions as you need, then search, filter, and resume them from the left sidebar. Session search covers titles, and you can search transcript contents too.

The session list filtered by the search term bug report, with a Search contents option and matching sessions listed with match counts — Search sessions by title, or use Search contents to find matches inside transcripts.

Useful session controls: * **Turn summary stats**: each agent turn shows file count and line changes (for example, "Finished in 6m 57s, 3 files +45 -12"). * **Click-to-copy code blocks**: click or tap inline code blocks in transcripts to copy them. * **Maximize button**: in the chat sidebar, open the current session in full Agent mode. * **Drag-drop to create workstreams**: drag a session onto a standalone session to group them. See [Workstreams](/session-management/workstreams.md). * **Start a session from a document**: "Start a new agent session" from a document pre-populates an @file reference to that document. * **Rename a session**: right-click a session in the list and choose Rename. Workstream parent rows also support inline rename. * **Preferred Agent Language**: under **Settings > Application > Agent Features**, set the language used to auto-generate session names. If you keep long-running conversations, Nimbalyst preserves and organizes history that the terminal CLI would lose.

A comparison table titled Nimbalyst unleashes Claude Code from the terminal, contrasting Nimbalyst and the Claude Code CLI on session management features — How Nimbalyst session management compares with running Claude Code in a terminal.

### Start a Background Session from Anywhere Press **Cmd/Ctrl+Shift+N** from any Nimbalyst mode to open the session launcher over your current work.

The Launch New Session composer floating over an open Excalidraw document, with agent, model, effort, and Actions controls — Start a session without leaving the document, Tracker, pull request, or shared file you are viewing.

The launcher includes the same controls as the normal chat composer: * Choose the agent and model. * Set effort and supported thinking controls. * Insert an Action or slash command. * Mention files and Tracker items. * Attach images and other supported context. Drag the launcher by its title bar when it covers something you need to read. Sending the prompt starts the session in the background, closes the launcher, and keeps you in the current mode. Open Agent mode or the attention list whenever you want to follow the new session. Starting from a document or Tracker item can prefill the relevant reference so the agent receives the context you launched from. ### Change Models and Thinking from Chat Press **Cmd/Ctrl+Shift+M** while the chat input is focused to open the model picker without reaching for the mouse. Type to search by the model's display name or model ID, then use the arrow keys and Enter to select it. The model selection belongs to the session. Changing it affects later turns without changing the model used by other sessions. Supported Claude Agent models also show **Extended: On** or **Extended: Off**: * **Extended: On** keeps the model's extended thinking behavior. * **Extended: Off** can reduce latency and token use when the task does not need as much reasoning. The control appears only for models and providers that support it. Extended thinking stays on by default. ### Review the Context Sent with a Prompt When you select content in an editor, the chat composer shows that selection as a removable chip before you send the prompt. Selection chips can represent: * Selected text in documents and code files * Spreadsheet cells * A selected mockup screen * One or more Excalidraw shapes * Selection types contributed by other editors Remove a chip when you do not want that selection sent to the model. Removing it changes only the pending prompt; it does not change or deselect the content in the editor. ### Contextual Tips Empty sessions and other core surfaces can show contextual tips instead of a blank state. These cards point you toward features that fit what you are doing right now, such as opening Tracker mode when your work spreads across many sessions, worktree flows, keyboard shortcuts, shared docs, and mobile pairing. Tips often include a direct action like **Open Tracker**, plus **Next** and **All tips** controls so you can browse more guidance without interrupting your work. ### Session Kanban Board Open the Session Kanban with **Cmd/Ctrl+Shift+K** to see all sessions in the workspace grouped by phase (Backlog, Planning, Implementing, Validating, Complete). Hover any session card and its transcript opens in a peek popover. If the session is running, the transcript streams in live, so you can watch what the agent is doing without opening the session. The peek shows the recent context of long turns, not just the trailing tokens, and works for sessions, workstreams, and child sessions inside a workstream. ### Self-Pacing Session Wakeups Agents can schedule themselves to wake up later and continue working, useful for long-running tasks where the agent needs to check back after a delay rather than running continuously. * The agent picks the wakeup time itself, anywhere from 60 seconds to 7 days out. * Wakeups persist across app restarts, but only fire while Nimbalyst is running. * When a wakeup fires, you get an OS notification; clicking it focuses the workspace and opens the session. * A wakeup banner appears in the session header with **Cancel** and **Fire now** controls, and sessions with a pending wakeup show a clock icon in the session list. If the workspace window is closed when the wakeup is due, the wakeup waits and re-fires the next time you open that workspace. ### Agent Attention List The Agent navigation icon shows a badge when sessions are running, unread, or waiting for you.

The Agent navigation icon with a 9+ badge next to a Sessions panel showing 166 need attention and an Awaiting Input group — The attention list groups sessions by what they need, so a blocked agent does not disappear into session history.

Open the icon to see sessions grouped as: * **Awaiting input**: the agent is blocked on a question, approval, or permission. * **Running**: the agent is actively working. * **Unread**: the session has new activity you have not opened. Select a session to open it. **Mark all read** clears unread state, but it does not approve requests or answer sessions that are awaiting input. Agents can also send a local system notification when a workflow needs your attention and you have stepped away. Clicking the notification returns you to the relevant workspace and session. Notification availability follows your operating-system and Nimbalyst notification settings. ### Menu Bar Session Fleet On macOS, the menu bar shows your session fleet without opening the app: * The menu bar names a session as it starts, finishes, blocks, or fails, and flags a session that has stopped responding. * When nothing is running, it quiets to a single mark. * Click the session name the menu bar is showing to open that session in Nimbalyst. * Open the panel to see which sessions are running, completed, or need attention, and click any session to navigate directly to it. * The sessions panel can mark every unread session as read at once. * A settings row in the panel lets you turn any of this behavior off. * The dock and Agent navigation badges show a count for sessions that need your attention. In session history, a session that launched other sessions is marked with an icon; hover it for a tooltip showing how many sessions it launched. ### Window State Persistence The Agent window remembers, per workspace: * Window size and position * Active tab in the sidebar * Session ID * Developer tools state (if open) * Transcript scroll position --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.nimbalyst.com/session-management/agent-window-and-session-management.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.