Create powerful collaborative AI workflows by connecting multiple MCP servers including Firecrawl Web Search, Exa, Monday for enhanced multi-agent automation capabilities in Klavis AI.
Advanced web crawling, scraping, and search capabilities powered by Firecrawl
Exa is an AI-powered search engine designed for AI applications. Use neural search to understand meaning and context, find similar content, get direct answers with citations, and conduct comprehensive research with structured analysis
Monday.com is a work operating system that powers teams to run projects and workflows with confidence. Create boards, manage items, customize columns, organize groups, and collaborate with team members in a visual workspace
Follow these steps to connect CrewAI to these MCP servers
Sign up for KlavisAI to access our MCP server management platform.
Set up your CrewAI agents with your desired MCP servers tools and configure authentication settings for collaborative workflows.
Test your multi-agent workflows and start using your enhanced collaborative AI capabilities.
import os
from crewai import Agent, Task, Crew, Process
from crewai_tools import MCPServerAdapter
from klavis import Klavis
from klavis.types import McpServerName, ConnectionType
# Initialize clients
klavis_client = Klavis(api_key=os.getenv("KLAVIS_API_KEY"))
firecrawl_web_search_mcp_instance = klavis_client.mcp_server.create_server_instance(
server_name=McpServerName.FIRECRAWL_WEB_SEARCH,
user_id="1234",
platform_name="Klavis",
connection_type=ConnectionType.STREAMABLE_HTTP,
)
exa_mcp_instance = klavis_client.mcp_server.create_server_instance(
server_name=McpServerName.EXA,
user_id="1234",
platform_name="Klavis",
connection_type=ConnectionType.STREAMABLE_HTTP,
)
monday_mcp_instance = klavis_client.mcp_server.create_server_instance(
server_name=McpServerName.MONDAY,
user_id="1234",
platform_name="Klavis",
connection_type=ConnectionType.STREAMABLE_HTTP,
)
# Initialize MCP tools for each server
firecrawl_web_search_tools = MCPServerAdapter(firecrawl_web_search_mcp_instance.server_params)
exa_tools = MCPServerAdapter(exa_mcp_instance.server_params)
monday_tools = MCPServerAdapter(monday_mcp_instance.server_params)
# Create specialized agents for each service
firecrawl_web_search_agent = Agent(
role="Firecrawl Web Search Specialist",
goal="Handle all Firecrawl Web Search related tasks and data processing",
backstory="You are an expert in Firecrawl Web Search operations and data analysis",
tools=firecrawl_web_search_tools,
reasoning=True,
verbose=False
)
exa_agent = Agent(
role="Exa Specialist",
goal="Handle all Exa related tasks and data processing",
backstory="You are an expert in Exa operations and data analysis",
tools=exa_tools,
reasoning=True,
verbose=False
)
monday_agent = Agent(
role="Monday Specialist",
goal="Handle all Monday related tasks and data processing",
backstory="You are an expert in Monday operations and data analysis",
tools=monday_tools,
reasoning=True,
verbose=False
)
# Define collaborative tasks
research_task = Task(
description="Gather comprehensive data from all available sources",
expected_output="Raw data and initial findings from all services",
agent=firecrawl_web_search_agent,
markdown=True
)
analysis_task = Task(
description="Analyze and synthesize the gathered data",
expected_output="Comprehensive analysis with insights and recommendations",
agent=exa_agent,
markdown=True
)
# Create multi-agent crew
multi_agent_crew = Crew(
agents=[firecrawl_web_search_agent, exa_agent, monday_agent],
tasks=[research_task, analysis_task],
verbose=False,
process=Process.sequential
)
result = multi_agent_crew.kickoff()
Everything you need to know about connecting CrewAI to these MCP servers
Join developers who are already using KlavisAI to power their CrewAI multi-agent systems with these MCP servers.
Start For Free