CrewAI Adapter

Generate Python CrewAI agent code from your agent.yaml manifest.

Usage

export ANTHROPIC_API_KEY=your-api-key-here
agentspec generate agent.yaml --framework crewai --output ./generated/

Get an API key at console.anthropic.com.

Generated Files

File	When generated
crew.py	Always — agent, task, and crew definitions
tools.py	When spec.tools is non-empty
guardrails.py	When spec.guardrails is set
requirements.txt	Always
.env.example	Always
README.md	Always

Manifest → Code Mapping

agent.yaml field	Generated code
spec.model.provider: groq	from langchain_groq import ChatGroq → Agent(llm=llm)
spec.model.provider: openai	from langchain_openai import ChatOpenAI
spec.model.provider: anthropic	from langchain_anthropic import ChatAnthropic
spec.model.parameters.temperature	temperature=N kwarg
spec.model.apiKey: $env:VAR	api_key=os.environ.get("VAR")
spec.prompts.system	backstory field on the Agent
spec.tools[]	@tool-decorated functions in tools.py, passed to Agent(tools=[...])
spec.memory	Crew(memory=True)
spec.guardrails.input[]	run_input_guardrails() in guardrails.py
spec.guardrails.output[]	run_output_guardrails() in guardrails.py
spec.observability.tracing.backend: langfuse	Langfuse env vars for automatic CrewAI tracing
spec.requires.envVars[]	validate_env() called at module top-level

crew.py Structure

# crew.py (excerpt)
from crewai import Agent, Task, Crew
from langchain_groq import ChatGroq

llm = ChatGroq(model="llama-3.3-70b-versatile", api_key=os.environ.get("GROQ_API_KEY"))

agent = Agent(
    role="GymCoach",
    goal="Help users achieve their fitness goals",
    backstory=load_system_prompt(),
    tools=[log_workout, get_workout_history],
    llm=llm,
    verbose=True,
)

def run_agent(user_input: str) -> str:
    task = Task(
        description=user_input,
        agent=agent,
        expected_output="A helpful, accurate response to the user's request.",
    )
    crew = Crew(agents=[agent], tasks=[task], memory=True, verbose=True)
    return str(crew.kickoff())

See also

• LangGraph adapter — Python LangGraph (multi-step graph execution)
• Mastra adapter — TypeScript Mastra
• Concepts: Adapters — how the generation system works