Skip to main content
from praisonaiagents import Agent
from langchain_community.tools import WikipediaQueryRun

agent = Agent(
    name="LangChain Wrapper",
    instructions="Use wrapped LangChain tools when helpful.",
    tools=[WikipediaQueryRun()],
)
agent.start("Summarise the Wikipedia article on transformer models")
The user asks a research question; the agent calls a LangChain tool wrapped for PraisonAI.

​
Langchain Tools

​
Quick Start

1

Install

pip install praisonaiagents langchain langchain-community
2

Wrap a LangChain tool

from praisonaiagents import Agent
from langchain_community.tools import WikipediaQueryRun
from langchain_community.utilities import WikipediaAPIWrapper

wiki_tool = WikipediaQueryRun(api_wrapper=WikipediaAPIWrapper())

def wikipedia_search(query: str) -> str:
    return wiki_tool.run(query)

agent = Agent(
    name="WikiAgent",
    instructions="Search Wikipedia for information.",
    tools=[wikipedia_search],
)

agent.start("What is quantum computing?")

​
Integrate Langchain Direct Tools

pip install youtube_search praisonai langchain_community langchain

# tools.py
from langchain_community.tools import YouTubeSearchTool

# agents.yaml
framework: crewai
topic: research about the causes of lung disease
agents:  # Canonical: use 'agents' instead of 'roles'
  research_analyst:
    instructions:  # Canonical: use 'instructions' instead of 'backstory' Experienced in analyzing scientific data related to respiratory health.
    goal: Analyze data on lung diseases
    role: Research Analyst
    tasks:
      data_analysis:
        description: Gather and analyze data on the causes and risk factors of lung
          diseases.
        expected_output: Report detailing key findings on lung disease causes.
    tools:
    - 'YouTubeSearchTool'

​
Integrate Langchain with Wrappers

pip install wikipedia langchain_community

# tools.py
from langchain_community.utilities import WikipediaAPIWrapper
class WikipediaSearchTool(BaseTool):
    name: str = "WikipediaSearchTool"
    description: str = "Search Wikipedia for relevant information based on a query."

    def _run(self, query: str):
        api_wrapper = WikipediaAPIWrapper(top_k_results=4, doc_content_chars_max=100)
        results = api_wrapper.load(query=query)
        return results

# agents.yaml
framework: crewai
topic: research about nvidia growth
agents:  # Canonical: use 'agents' instead of 'roles'
  data_collector:
    instructions:  # Canonical: use 'instructions' instead of 'backstory' An experienced researcher with the ability to efficiently collect and
      organize vast amounts of data.
    goal: Gather information on Nvidia's growth by providing the Ticket Symbol to YahooFinanceNewsTool
    role: Data Collector
    tasks:
      data_collection_task:
        description: Collect data on Nvidia's growth from various sources such as
          financial reports, news articles, and company announcements.
        expected_output: A comprehensive document detailing data points on Nvidia's
          growth over the years.
    tools:
    - 'WikipediaSearchTool'

​
Best Practices

Convert LangChain tools to plain Python functions with tool.run() for cleaner integration.
Prefer PraisonAI native tools for search and memory - use LangChain only when there is no native equivalent.
LangChain tools can raise ToolException - wrap calls in try/except and return error strings.
Tools with str-to-str signatures integrate most reliably with PraisonAI agents.

Custom Tools

Build your own agent tools

Tools Overview

Browse PraisonAI tool documentation