CrewAI Integration Guide

pip install crewai
# For additional tools (optional)
pip install 'crewai[tools]'

import os
from crewai import Agent, Task, Crew, LLM

# Set up with environment variables
os.environ["OPENAI_API_KEY"] = "your-apipie-api-key"
os.environ["OPENAI_API_BASE"] = "https://apipie.ai/v1"

# Or configure directly in your code
llm = LLM(
    provider="openai",
    api_key="your-apipie-api-key",
    base_url="https://apipie.ai/v1",
    model="gpt-4o-mini"  # Or any model available on APIpie
)

import os
from crewai import Agent, Task, Crew, LLM

# Configure APIpie as the LLM provider
os.environ["OPENAI_API_KEY"] = "your-apipie-api-key"
os.environ["OPENAI_API_BASE"] = "https://apipie.ai/v1"

# Alternatively, configure the LLM explicitly
llm = LLM(
    provider="openai",
    api_key="your-apipie-api-key",
    base_url="https://apipie.ai/v1",
    model="gpt-4o-mini"  # Or any model available on APIpie
)

# Create a researcher agent
researcher = Agent(
    role="Research Analyst",
    goal="Find and summarize information about specific topics",
    backstory="You are an experienced researcher with attention to detail",
    llm=llm,  # Use the configured LLM
    verbose=True
)

# Create a writer agent
writer = Agent(
    role="Content Writer",
    goal="Create engaging and informative content from research",
    backstory="You are a skilled writer who turns complex information into readable content",
    llm=llm,  # Use the configured LLM
    verbose=True
)

# Create tasks for the agents
research_task = Task(
    description="Conduct thorough research about artificial intelligence trends in 2025",
    expected_output="A list with 10 bullet points of the most relevant AI trends in 2025",
    agent=researcher
)

writing_task = Task(
    description="Write a comprehensive article about AI trends in 2025 based on the research provided",
    expected_output="A 1000-word article with clear sections covering each major trend",
    agent=writer
)

# Create a crew with sequential process
crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, writing_task],
    verbose=True
)

# Execute the crew
result = crew.kickoff()
print(result)

import os
from crewai import Agent, Task, Crew, LLM
from crewai_tools import SerperDevTool  # Search tool

# Configure your credentials
os.environ["OPENAI_API_KEY"] = "your-apipie-api-key"
os.environ["OPENAI_API_BASE"] = "https://apipie.ai/v1"
os.environ["SERPER_API_KEY"] = "your-serper-api-key"  # For the search tool

# Create a researcher agent with search capabilities
researcher = Agent(
    role="Market Researcher",
    goal="Research market trends and competitor analysis",
    backstory="You're a skilled market researcher with years of experience in competitor analysis",
    tools=[SerperDevTool()],  # Add the search tool
    llm=LLM(
        provider="openai",
        api_key=os.environ["OPENAI_API_KEY"],
        base_url=os.environ["OPENAI_API_BASE"],
        model="gpt-4o"
    ),
    verbose=True
)

# Create a financial analyst agent
analyst = Agent(
    role="Financial Analyst",
    goal="Analyze financial data and provide investment recommendations",
    backstory="You're a financial expert who excels at interpreting market data",
    llm=LLM(
        provider="openai",
        api_key=os.environ["OPENAI_API_KEY"],
        base_url=os.environ["OPENAI_API_BASE"],
        model="gpt-4o"
    ),
    verbose=True
)

# Define tasks
research_task = Task(
    description="Research the current state of AI chip manufacturing. Focus on major players, market share, and recent innovations.",
    expected_output="A detailed report on the AI chip market landscape with key players and market trends",
    agent=researcher
)

analysis_task = Task(
    description="Based on the research, analyze the investment potential of top AI chip manufacturers. Consider growth prospects, risks, and competitive advantages.",
    expected_output="An investment analysis report with recommendations for the top 3 AI chip manufacturers",
    agent=analyst
)

# Create and execute the crew
crew = Crew(
    agents=[researcher, analyst],
    tasks=[research_task, analysis_task],
    verbose=True
)

result = crew.kickoff()
print(result)

import os
from crewai import Agent, Task, Crew, LLM
from crewai.flow.flow import Flow, listen, start, router
from pydantic import BaseModel

# Configure APIpie
os.environ["OPENAI_API_KEY"] = "your-apipie-api-key"
os.environ["OPENAI_API_BASE"] = "https://apipie.ai/v1"

# Define a state model for the flow
class ResearchState(BaseModel):
    topic: str = ""
    confidence: float = 0.0
    findings: list = []

# Create a flow with structured control
class ResearchFlow(Flow[ResearchState]):
    @start()
    def initialize_research(self):
        self.state.topic = "Machine Learning Applications in Healthcare"
        return {"topic": self.state.topic}

    @listen(initialize_research)
    def conduct_research(self, inputs):
        # Create a research agent
        researcher = Agent(
            role="Medical Research Specialist",
            goal="Find and analyze the latest applications of ML in healthcare",
            backstory="You're a specialist in healthcare technology research",
            llm=LLM(
                provider="openai",
                api_key=os.environ["OPENAI_API_KEY"],
                base_url=os.environ["OPENAI_API_BASE"],
                model="gpt-4o-mini"
            )
        )

        # Create a research task
        research_task = Task(
            description=f"Research the latest applications of machine learning in healthcare, focusing on {inputs['topic']}",
            expected_output="A comprehensive report on ML applications in healthcare",
            agent=researcher
        )

        # Create and execute a simple crew
        research_crew = Crew(
            agents=[researcher],
            tasks=[research_task],
            verbose=True
        )

        result = research_crew.kickoff()

        # Update state with findings
        self.state.findings.append(result)
        self.state.confidence = 0.75

        return result

    @router(conduct_research)
    def determine_next_step(self):
        # Conditional routing based on confidence
        if self.state.confidence > 0.7:
            return "sufficient_data"
        else:
            return "need_more_data"

    @listen("sufficient_data")
    def generate_final_report(self):
        # Create a report writing agent
        writer = Agent(
            role="Medical Content Writer",
            goal="Transform research into readable reports",
            backstory="You specialize in making complex medical technology understandable",
            llm=LLM(
                provider="openai",
                api_key=os.environ["OPENAI_API_KEY"],
                base_url=os.environ["OPENAI_API_BASE"],
                model="gpt-4o-mini"
            )
        )

        # Create a report writing task
        report_task = Task(
            description="Transform the research findings into a comprehensive report for healthcare professionals",
            expected_output="A detailed report on ML applications in healthcare with recommendations",
            agent=writer,
            context="\n".join(self.state.findings)
        )

        # Create and execute the crew
        report_crew = Crew(
            agents=[writer],
            tasks=[report_task],
            verbose=True
        )

        return report_crew.kickoff()

    @listen("need_more_data")
    def request_additional_research(self):
        # Logic for when more research is needed
        return "More research needed to increase confidence in findings"

# Execute the flow
research_flow = ResearchFlow()
final_result = research_flow.start()
print(final_result)

import os
from crewai import Agent, Task, Crew, LLM, Process
from crewai.project import CrewBase, agent, crew, task
from crewai_tools import SerperDevTool
from crewai.agents.agent_builder.base_agent import BaseAgent
from typing import List

# Configure APIpie
os.environ["OPENAI_API_KEY"] = "your-apipie-api-key"
os.environ["OPENAI_API_BASE"] = "https://apipie.ai/v1"

@CrewBase
class ResearchCrew():
    """Research crew for analyzing topics"""
    agents: List[BaseAgent]
    tasks: List[Task]

    @agent
    def researcher(self) -> Agent:
        return Agent(
            config=self.agents_config['researcher'],
            verbose=True,
            tools=[SerperDevTool()],
            llm=LLM(
                provider="openai",
                api_key=os.environ["OPENAI_API_KEY"],
                base_url=os.environ["OPENAI_API_BASE"],
                model="gpt-4o-mini"
            )
        )

    @agent
    def analyst(self) -> Agent:
        return Agent(
            config=self.agents_config['analyst'],
            verbose=True,
            llm=LLM(
                provider="openai",
                api_key=os.environ["OPENAI_API_KEY"],
                base_url=os.environ["OPENAI_API_BASE"],
                model="gpt-4o-mini"
            )
        )

    @task
    def research_task(self) -> Task:
        return Task(
            config=self.tasks_config['research_task'],
        )

    @task
    def analysis_task(self) -> Task:
        return Task(
            config=self.tasks_config['analysis_task'],
            output_file='analysis.md'
        )

    @crew
    def crew(self) -> Crew:
        """Creates the research crew"""
        return Crew(
            agents=self.agents,
            tasks=self.tasks,
            process=Process.sequential,
            verbose=True,
        )

# Execute the crew with inputs
inputs = {
    'topic': 'Quantum Computing Applications'
}
result = ResearchCrew().crew().kickoff(inputs=inputs)
print(result)