langchain - 💡(How to fix) Fix Add Disvr tool discovery integration to langchain-community [3 comments, 3 participants]

Fix Plan

To integrate Disvr into langchain-community, we will create a new package, langchain-disvr, and add two main functions: DisvrDiscoverTool and DisvrToolDetails.

Implementation Steps

• Install the langchain-disvr package using pip: pip install langchain-disvr
• Import the package in your langchain-community code: import langchain_disvr
• Use the DisvrDiscoverTool function to find the best tools for a task:

from langchain_disvr import DisvrDiscoverTool

def find_best_tool(task):
    disvr_tool = DisvrDiscoverTool()
    best_tool = disvr_tool.find_best_tool(task)
    return best_tool

• Use the DisvrToolDetails function to get install instructions, env vars, and tool functions:

from langchain_disvr import DisvrToolDetails

def get_tool_details(tool_id):
    disvr_tool = DisvrToolDetails()
    tool_details = disvr_tool.get_tool_details(tool_id)
    return tool_details

Verification

To verify the integration, you can test the find_best_tool and get_tool_details functions with sample inputs:

task = "text classification"
best_tool = find_best_tool(task)
print(best_tool)

tool_id = best_tool["id"]
tool_details = get_tool_details(tool_id)
print(tool_details)

This should output the best tool for the given task and its details.

Extra Tips

• Make sure to handle errors and exceptions properly when using the DisvrDiscoverTool and DisvrToolDetails functions.
• You can customize the DisvrDiscoverTool function to use different ranking algorithms or filters.
• Refer to the Disvr documentation for more information on how to use the API.