Introduction

This section will introduce the deployment process. More application scenarios will be supported in the future.

General Bot Configuration

1. A solution is provided for handling images returned in base64 format when the bot sends images externally. You can add an image hosting service in the general bot settings. More image hosting options will be available in the future; currently, only EasyImage2 is supported.

QQ Official Bot

This update introduces support for QQ bots. You can now link your configured agent to a QQ official bot without the risk of account suspension. To use this feature:

• Register a bot on the QQ Open Platform and enter the App ID and App Secret in the corresponding fields.
• Click the button in the top-right corner to activate the bot.

QQ Bot Features

1. Customizable Message Delimiters: By default, the delimiters are ["。", "\\\\n", "？", "！"]. The bot streams the agent's output and sends it to QQ line by line based on these delimiters.
2. Supports Private and Group Chats: The QQ bot works in both private messages and group conversations.
3. Text and Image Messages Supported: It supports sending and receiving text and image messages. If the model has vision capabilities or a vision model is enabled, it can handle images. You can also enable a text-to-image model to generate and send images.
4. Adjustable Conversation Rounds: The default is 30 rounds. Older conversation records are gradually discarded to prevent context overflow. You can configure a memory module to maintain character personality and achieve permanent memory retention.
5. Displays Thinking and Tool Call Process: The QQ bot can return the agent’s thinking process and tool usage steps. You can choose whether to enable this feature in the configuration.

Deployment steps

1. Log in to QQ Open Platform and register a robot. Each person can create up to 4 robots.

1. Copy the robot ID and key to the agent party.