AI Chat with the awork Agent

The awork Agent knows your awork context and can do more than just answer questions - it can also look things up in awork and perform actions there.

---

Main Benefits and Use Cases

The awork Agent is the built-in agent. You can get started right away without setting up an agent yourself.

It helps you with things like:

• Finding information - Locate projects, tasks, or other information in awork

• Getting things done - Create tasks or modify existing content

• Using the web and images - The agent can perform web searches and generate images

• Working with your awork context - Link projects, documents, and files directly to a thread

• Using voice input - Dictate your input through the microphone instead of typing

Typical questions might be about the current project status, available capacity, or assigning a task to a customer.

---

How to Open the Chat Interface

You can access the awork Agent in 3 ways:

• Click on AI in the left main menu
• Click the awork AI chat button in the floating action bar at the bottom of the screen (recognizable by the robot icon)
• Use the shortcut Cmd/Ctrl + I to open the AI chat sidebar

The views share the same thread history.

A text input field for your prompt will appear.

The Input Area

The input area is the central space for interacting with an agent.

View Threads
Click the Threads icon in the panel header to open an overview of your previous conversations. Threads are organized by categories like "Waiting for Action" and "Recently Used".

Customize the View
Various view options are available in the chat window header:

• Show as Sidebar: Docks the panel as a sidebar on the right edge

• Show as Popup: Displays the panel as a floating window

• Expand Size: Enlarges the panel

• Reduce Size: Shrinks the panel

• Dock and Undock Sidebar
  When the panel is displayed as a sidebar, you can use Dock AI Chat Sidebar to pin it in place or use Undock AI Chat Sidebar to release it.

• Close Chat Panel
  Click the X icon in the top right of the panel header to close the chat window. Alternatively, you can press the Esc key.

---

Managing AI Threads

A conversation with an agent is called a Thread.

Your threads are saved so you can return to them later.

The Thread List on the left gives you access to recently used threads.

Here you can:

• Create threads

• Search for threads

• Pin threads

• Open and reuse threads

Thread Details in the top right show information such as:

• which agent was used
• where the thread was started
• which project or task was involved
• whether it was started by an automation
• whether it was scheduled
• who started it
• when it was started

The corresponding project or task can be opened directly from the thread details.

The Action Button for a thread or the top right offers these options:

• Rename (thread names are auto-generated)

• Regenerate Title

• Add to Project

• Copy ID

• Copy Link

• Archive: Archived threads are moved to an archive area, can't be actively edited further, but can be restored later

• Delete

A thread contains more than just its messages. This includes files generated by the agent as well as tools and connectors that were used.

Thread Status

AI threads can display different statuses and indicators.

• Running: The agent is currently working on a request. A Loading Indicator is shown during processing.

• Done: The agent has completed the current processing. A purple dot appears to the left of the thread title.

• Agent Waiting for Your Input: When an agent asks a question and needs a decision or input, the AI thread is marked with a speech bubble icon. Such AI threads are temporarily grouped in the Waiting for Action category, so you can quickly see where your input is needed. Depending on the request, the agent might display a selection or a custom input form. After you provide the input, the agent can continue processing.

Unread Thread Messages

If a new message arrives in an AI thread since your last visit, you'll see it marked by a notification counter on the AI menu item in the left main menu. This is especially helpful for scheduled or automatic agent runs.

---

Selecting and Using an Agent

Every new thread starts with the awork Agent by default.

You can instead select a custom agent when creating a new thread:

1. Open a new thread

2. Click on awork Agent and select a custom agent from the dropdown

3. Start the thread

The selection shows available custom agents alphabetically. There's no list for recently used or favorite agents.

After the first message, the agent for that thread is locked in. You can't switch agents within a thread. To use a different agent, you need to start a new thread.

Custom agent threads show the respective agent's avatar in the thread list. Threads with the awork Agent don't show an agent avatar.

---

Working with Context from awork

The awork Agent can work with your existing awork context.

You can enter text in the message input area, but you can also add context directly.

For example, if a project is selected through the context menu, the AI receives the exact object ID. This prevents confusion with objects that have the same name.

• Use the @ mention to reference awork objects and include projects, documents, or files in a thread.

• Or click the Plus Icon at the bottom left of the input field. A menu labeled "Context & Capabilities" will appear with various options for adding content.

  • Upload File: Add files as context. Alternatively, you can insert files via drag and drop or copy and paste

  • Add Context: Provide additional context.
    Choose Person, Team, Customer, Task, Project, Doc, or Current Page Context.

Context doesn't necessarily mean the entire content of a referenced object is always fully inserted into the prompt.

Instead, the AI evaluates the available context together with your intent and decides which information is relevant.

This lets you reference projects, documents, and other objects without having to manually copy their entire content into your prompt.

Uploading Files

You can attach files.

These include:

• Images
• PDFs
• Other documents

You can also insert links.

Using awork AI from Anywhere

You can open the AI from different areas of awork.

The currently open page can automatically become part of the context for a new AI interaction. For example, you can open the AI while working on a project or document and ask questions directly about that page.

---

Adding Capabilities

Click the Plus Icon at the bottom left of the input field. In the opened menu, you'll find these options:

• Skills: You can explicitly add a skill to your AI thread or reference it in your prompt. In the panel that appears, you can manage more skills in the library. The AI evaluates whether the skill is relevant for your request.

  If a skill is explicitly mentioned, it's more likely to be used because you've directly referred to it.

• Connectors (MCP): Set up external connections. Available connectors are shown in the side panel, which you can enable or disable using the toggle switches.

• Tools: Enable or disable tools like web search and awork access. Web access is recommended for current information. awork access should be enabled if you want to query or work with information from awork. These tools can still be deliberately disabled if you want to limit available capabilities or reduce consumption.

---

Controlling the awork Agent's Thinking Mode

Choose Performance Mode
You can select how fast or thorough you want the awork Agent to work. Click the Balanced button below the input field.

Three simple presets are available:

• Fast - For quick answers

• Balanced - Balanced performance (default setting)

• Deep thinking - For complex tasks requiring more thinking

Enable Advanced Settings
Below the presets, you'll find the Advanced option with a toggle switch. Enable this to display additional settings like Reasoning Level, Image Model, and a selection of different AI Models.

This lets you override the preset and select the underlying model and reasoning level yourself.

The reasoning level determines how much effort the model puts in before answering. A higher level can produce better results for difficult tasks, but works more slowly.

In the workspace settings, you can determine which model and reasoning level are used behind the three presets. Which models are available also depends on your central workspace settings.

Switching Models During a Thread

You can also switch the model after starting a thread.

Some models use technical prompt caching during inference to more efficiently reuse already processed context. No prompt or output content is persistently stored in this process.

Different models may interpret the same conversation context differently, which can affect the consistency of responses. Additionally, switching models can lead to higher credit consumption because technical prompt caching can't be reused across models.

---

Sending Messages During Processing

You don't need to wait for the agent to finish answering.

If you send another message while a run is in progress, it gets queued. Up to 10 messages can be processed sequentially per thread.

Processing happens in the order you sent the messages by default.

You can also prioritize a message that's already waiting. This will end the current run and start the selected message instead.

---

Details of an Agent Run

While an agent is working, you can click on a dropdown arrow to expand and collapse a section showing reasoning and activity information.

The display distinguishes between:

• Reasoning or intermediate activities, which are displayed in a muted style
• Final answers, which are highlighted as the actual response

The reasoning section might contain information about:

• Searches
• Tool calls
• Individual processing steps
• Intermediate thoughts during execution

Each message also includes a timestamp and a button to copy the response.

---

Using Voice Input

There's a microphone icon in the input field. You can use it to dictate your input. Your spoken prompt will be converted to text and sent as a normal message to the agent.

---

Performing Actions with the awork Agent

The awork Agent can do more than just read awork data - it can also edit it.

This includes things like creating or modifying tasks and documents.

Before the agent creates, edits, or deletes anything in awork, it asks for your confirmation.

---

Memory and Workspace Context

The awork Agent can store information about you, a customer, or a project.

Different rules apply:

• Personal information applies to you personally

• Project-related information is used when the agent works in that project context

• Customer-related information is used when the agent works in the corresponding customer context

This memory runs in the background. There's no view where you can see or manage stored memories.

In addition, there's the Workspace Context. Your team can store shared background knowledge there, such as company knowledge, terms, rules, and ways of working.

All workspace members can read the workspace context. Only users with the appropriate workspace configuration permissions can edit it.

The workspace context doesn't give agents additional tools or permissions.

---

FAQ

Is the awork Agent a custom agent?

No. The awork Agent is the built-in agent for all users. Custom agents, on the other hand, are saved agents that can be set up and shared for specific recurring tasks.

Can I switch agents within a thread?

No. Once the first message is sent, the agent for that thread is locked in. To use a different agent, you need to start a new thread.

Can the awork Agent make changes in awork?

Yes. For example, it can create or modify tasks and documents. Before write actions in awork, it asks for your confirmation.

Can the awork Agent use other tools?

Yes. It can search and edit awork data, use web search, generate images, and access other tools through connected apps.

Can I see what the agent did?

Yes. Agent actions are logged in awork's activity logs with the responsible user and the agent used, making them fully traceable.