> ## Documentation Index
> Fetch the complete documentation index at: https://phidatainc-agui.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Discord Bot

> Host agents as Discord Bots.

The Discord Bot integration allows you to serve Agents or Teams via Discord, using the discord.py library to handle Discord events and send messages.

## Setup Steps

<Snippet file="setup-discord-app.mdx" />

### Example Usage

Create an agent, wrap it with `DiscordClient`, and run it:

```python theme={null}
from agno.agent import Agent
from agno.integrations.discord import DiscordClient
from agno.models.openai import OpenAIResponses

basic_agent = Agent(
    name="Basic Agent",
    model=OpenAIResponses(id="gpt-5.2"), 
    add_history_to_context=True,
    num_history_runs=3,
    add_datetime_to_context=True,
)

discord_agent = DiscordClient(basic_agent)
if __name__ == "__main__":
    discord_agent.serve()
```

## Core Components

* `DiscordClient`: Wraps Agno agents/teams for Discord integration using discord.py.
* `DiscordClient.serve`: Starts the Discord bot client with the provided token.

## `DiscordClient` Class

Main entry point for Agno Discord bot applications.

### Initialization Parameters

| Parameter | Type              | Default | Description            |
| --------- | ----------------- | ------- | ---------------------- |
| `agent`   | `Optional[Agent]` | `None`  | Agno `Agent` instance. |
| `team`    | `Optional[Team]`  | `None`  | Agno `Team` instance.  |

*Provide `agent` or `team`, not both.*

## Event Handling

The Discord bot automatically handles various Discord events:

### Message Events

* **Description**: Processes all incoming messages from users
* **Media Support**: Handles images, videos, audio files, and documents
* **Threading**: Automatically creates threads for conversations
* **Features**:
  * Automatic thread creation for each conversation
  * Media processing and forwarding to agents
  * Message splitting for responses longer than 1500 characters
  * Support for reasoning content display
  * Context enrichment with username and message URL

### Supported Media Types

* **Images**: Direct URL processing for image analysis
* **Videos**: Downloads and processes video content
* **Audio**: URL-based audio processing
* **Files**: Downloads and processes document attachments

## Environment Variables

Ensure the following environment variable is set:

```bash theme={null}
export DISCORD_BOT_TOKEN="your-discord-bot-token"
```

## Message Processing

The bot processes messages with the following workflow:

1. **Message Reception**: Receives messages from Discord channels
2. **Media Processing**: Downloads and processes any attached media
3. **Thread Management**: Creates or uses existing threads for conversations
4. **Agent/Team Execution**: Forwards the message and media to the configured agent or team
5. **Response Handling**: Sends the response back to Discord, splitting long messages if necessary
6. **Reasoning Display**: Shows reasoning content in italics if available

## Features

### Automatic Thread Creation

* Creates a new thread for each user's first message
* Maintains conversation context within threads
* Uses the format: `{username}'s thread`

### Media Support

* **Images**: Passed as `Image` objects with URLs
* **Videos**: Downloaded and passed as `Video` objects with content
* **Audio**: Passed as `Audio` objects with URLs
* **Files**: Downloaded and passed as `File` objects with content

### Message Formatting

* Long messages (>1500 characters) are automatically split
* Reasoning content is displayed in italics
* Batch numbering for split messages: `[1/3] message content`

## Testing the Integration

1. Set up your Discord bot token: `export DISCORD_BOT_TOKEN="your-token"`
2. Run your application: `python your_discord_bot.py`
3. Invite the bot to your Discord server
4. Send a message in any channel where the bot has access
5. The bot will automatically create a thread and respond