Discord
This notebook provides a quick overview for getting started with Discord tooling inlangchain_discord. For more details on each tool and configuration, see the docstrings in your repository or relevant doc pages.
Overview
Integration details
| Class | Package | Serializable | JS support | Package latest |
|---|---|---|---|---|
DiscordReadMessages,DiscordSendMessage | langchain-discord-shikenso | N/A | TBD |
Tool features
DiscordReadMessages: Reads messages from a specified channel.DiscordSendMessage: Sends messages to a specified channel.
Setup
The integration is provided by thelangchain-discord-shikenso package. Install it as follows:
%pip install--quiet-U langchain-discord-shikenso
Credentials
This integration requires you to setDISCORD_BOT_TOKEN as an environment variable to authenticate with the Discord API.
export DISCORD_BOT_TOKEN="your-bot-token"
import getpass
import os
# Example prompt to set your token if not already set:
# if not os.environ.get("DISCORD_BOT_TOKEN"):
# os.environ["DISCORD_BOT_TOKEN"] = getpass.getpass("DISCORD Bot Token:\n")
You can optionally set upLangSmith for tracing or observability:
# os.environ["LANGSMITH_TRACING"] = "true"
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass()
Instantiation
Below is an example showing how to instantiate the Discord tools inlangchain_discord. Adjust as needed for your specific usage.
from langchain_discord.tools.discord_read_messagesimport DiscordReadMessages
from langchain_discord.tools.discord_send_messagesimport DiscordSendMessage
read_tool= DiscordReadMessages()
send_tool= DiscordSendMessage()
# Example usage:
# response = read_tool({"channel_id": "1234567890", "limit": 5})
# print(response)
#
# send_result = send_tool({"message": "Hello from notebook!", "channel_id": "1234567890"})
# print(send_result)
Invocation
Direct invocation with args
Below is a simple example of calling the tool with keyword arguments in a dictionary.
invocation_args={"channel_id":"1234567890","limit":3}
response= read_tool(invocation_args)
response
Invocation with ToolCall
If you have a model-generatedToolCall, pass it totool.invoke() in the format shown below.
tool_call={
"args":{"channel_id":"1234567890","limit":2},
"id":"1",
"name": read_tool.name,
"type":"tool_call",
}
tool.invoke(tool_call)
Chaining
Below is a more complete example showing how you might integrate theDiscordReadMessages andDiscordSendMessage tools in a chain or agent with an LLM. This example assumes you have a function (likecreate_react_agent) that sets up a LangChain-style agent capable of calling tools when appropriate.
# Example: Using Discord Tools in an Agent
from langgraph.prebuiltimport create_react_agent
from langchain_discord.tools.discord_read_messagesimport DiscordReadMessages
from langchain_discord.tools.discord_send_messagesimport DiscordSendMessage
# 1. Instantiate or configure your language model
# (Replace with your actual LLM, e.g., ChatOpenAI(temperature=0))
llm=...
# 2. Create instances of the Discord tools
read_tool= DiscordReadMessages()
send_tool= DiscordSendMessage()
# 3. Build an agent that has access to these tools
agent_executor= create_react_agent(llm,[read_tool, send_tool])
# 4. Formulate a user query that may invoke one or both tools
example_query="Please read the last 5 messages in channel 1234567890"
# 5. Execute the agent in streaming mode (or however your code is structured)
events= agent_executor.stream(
{"messages":[("user", example_query)]},
stream_mode="values",
)
# 6. Print out the model's responses (and any tool outputs) as they arrive
for eventin events:
event["messages"][-1].pretty_print()
API reference
See the docstrings in:
for usage details, parameters, and advanced configurations.
Related
- Toolconceptual guide
- Toolhow-to guides