Skip to main content
Open In ColabOpen on GitHub

Discord

此笔记本为快速了解如何在Discord工具中入门提供了一个概览。有关每个工具和配置的更多详细信息，请参阅您仓库中的文档字符串或相关文档页面。

概览​

集成细节​

Class包序列化JS支持Package 最新版本
DiscordReadMessages, DiscordSendMessagelangchain-discord-shikensoN/ATBDPyPI - Version

Tool features​

  • DiscordReadMessages: 从指定频道读取消息。
  • DiscordSendMessage: 向指定的频道发送消息。

设置​

The integration is provided by the langchain-discord-shikenso package. Install it as follows:

%pip install --quiet -U langchain-discord-shikenso

Credentials​

此集成要求您将DISCORD_BOT_TOKEN设置为环境变量以通过Discord API进行身份验证。

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")

您可以选择设置 LangSmith 以进行跟踪或可观测性：

# os.environ["LANGSMITH_TRACING"] = "true"
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass()

Instantiation​

以下是一个示例，展示如何在langchain_discord中实例化Discord工具。根据具体需求进行相应调整。

from langchain_discord.tools.discord_read_messages import DiscordReadMessages
from langchain_discord.tools.discord_send_messages import 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​

直接调用带参数​

以下是使用字典作为关键字参数调用工具的一个简单示例。

invocation_args = {"channel_id": "1234567890", "limit": 3}
response = read_tool(invocation_args)
response

调用带有ToolCall​

如果您生成了一个模型输出ToolCall，请将其按照以下格式传递给tool.invoke()。

tool_call = {
    "args": {"channel_id": "1234567890", "limit": 2},
    "id": "1",
    "name": read_tool.name,
    "type": "tool_call",
}

tool.invoke(tool_call)

链式调用​

以下是一个更完整的示例，展示了你如何在LLM中将DiscordReadMessages和DiscordSendMessage工具集成到一个链或代理中。此示例假设您有一个函数（如create_react_agent），该函数能够设置一种类似于LangChain风格的代理，在适当的时候调用工具。

# Example: Using Discord Tools in an Agent

from langgraph.prebuilt import create_react_agent
from langchain_discord.tools.discord_read_messages import DiscordReadMessages
from langchain_discord.tools.discord_send_messages import 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 event in events:
    event["messages"][-1].pretty_print()

API 参考​

请参阅以下位置的文档字符串：

for usage details, parameters, and advanced configurations.