TaskWeaver User Interface: Create Code-First AI Agents with NO CODE
> [!info] AI Summary
>
> ### Task Viewer User Interface Summary
>
> - Introduction to Task Viewer UI
> - Excitement about showcasing the Task Viewer UI.
> - Step-by-step guide on installation and usage.
> - Reference to a previous video on Task Weaver framework.
>
> - Installation Steps
> 1. Clone the repository: `git clone Microsoft/TaskWeaver`.
> 2. Navigate to the cloned folder.
> 3. Create a new environment: `conda create -n taskViewer python=3.11`.
> 4. Activate the environment: `conda activate taskViewer`.
> 5. Install requirements: `pip install -r requirements.txt`.
> 6. Install Streamlit for UI: `pip install streamlit`.
>
> - Configuration and Setup
> - Open the folder in VS Code to view the structure.
> - Configure `taskWeaver config.js` with API base, key, and model name.
> - Obtain API key from OpenAI and save the file.
> - Export OpenAI API key in the terminal if necessary.
>
> - Running the Application
> - Navigate to the `Playground/ui` folder.
> - Start the app with `streamlit run app.py`.
> - Access the UI through the provided URL.
>
> - Using the User Interface
> - Assign tasks to agents without coding via the UI.
> - Example task: Plotting Apple stock price chart.
> - Ability to specify time range for data.
> - New chats and history viewing options available.
> - Capability to handle both basic and advanced tasks.
>
> ---