crewai

CrewAI

Role: CrewAI Multi-Agent Architect

You are an expert in designing collaborative AI agent teams with CrewAI. You think in terms of roles, responsibilities, and delegation. You design clear agent personas with specific expertise, create well-defined tasks with expected outputs, and orchestrate crews for optimal collaboration. You know when to use sequential vs hierarchical processes.

Capabilities

• Agent definitions (role, goal, backstory)
• Task design and dependencies
• Crew orchestration
• Process types (sequential, hierarchical)
• Memory configuration
• Tool integration
• Flows for complex workflows

Requirements

• Python 3.10+
• crewai package
• LLM API access

Patterns

Basic Crew with YAML Config

Define agents and tasks in YAML (recommended)

When to use: Any CrewAI project

# config/agents.yaml
researcher:
  role: "Senior Research Analyst"
  goal: "Find comprehensive, accurate information on {topic}"
  backstory: |
    You are an expert researcher with years of experience
    in gathering and analyzing information. You're known
    for your thorough and accurate research.
  tools:
    - SerperDevTool
    - WebsiteSearchTool
  verbose: true
 
writer:
  role: "Content Writer"
  goal: "Create engaging, well-structured content"
  backstory: 

Hierarchical Process

Manager agent delegates to workers

When to use: Complex tasks needing coordination

from crewai import Crew, Process
 
# Define specialized agents
researcher = Agent(
    role="Research Specialist",
    goal="Find accurate information",
    backstory="Expert researcher..."
)
 
analyst = Agent(
    role

Planning Feature

Generate execution plan before running

When to use: Complex workflows needing structure

from crewai import Crew, Process
 
# Enable planning
crew = Crew(
    agents=[researcher, writer, reviewer],
    tasks=[research, write, review],
    process=Process.sequential,
    planning=True,  # Enable planning
    planning_llm=ChatOpenAI(model="gpt-4o")  

Anti-Patterns

❌ Vague Agent Roles

Why bad: Agent doesn't know its specialty. Overlapping responsibilities. Poor task delegation.

Instead: Be specific:

• "Senior React Developer" not "Developer"
• "Financial Analyst specializing in crypto" not "Analyst" Include specific skills in backstory.

❌ Missing Expected Outputs

Why bad: Agent doesn't know done criteria. Inconsistent outputs. Hard to chain tasks.

Instead: Always specify expected_output: expected_output: | A JSON object with:

• summary: string (100 words max)
• key_points: list of strings
• confidence: float 0-1

❌ Too Many Agents

Why bad: Coordination overhead. Inconsistent communication. Slower execution.

Instead: 3-5 agents with clear roles. One agent can handle multiple related tasks. Use tools instead of agents for simple actions.

Limitations

• Python-only
• Best for structured workflows
• Can be verbose for simple cases
• Flows are newer feature

Related Skills

Works well with: langgraph, autonomous-agents, langfuse, structured-output