Skip to main content

What We’ll Build

A research crew that:
  1. Researches a topic using web search
  2. Analyzes and synthesizes findings
  3. Produces a written report as an artifact

Prerequisites

  • Python 3.11+
  • CrewAI installed (pip install crewai)
  • Crewship CLI installed and authenticated
Want to skip the setup? Clone the crewai-quickstart repo and run crewship deploy to get a working crew deployed in minutes.

Step 1: Create the Project

This creates:

Step 2: Define Your Agents

Edit src/research_crew/config/agents.yaml:
agents.yaml

Step 3: Define Your Tasks

Edit src/research_crew/config/tasks.yaml:
tasks.yaml
Files written to artifacts/ are collected as run artifacts, accessible via the API and Console.

Step 4: Configure the Crew

Edit src/research_crew/crew.py:
crew.py

Step 5: Create the Entry Point

Edit src/research_crew/main.py:
main.py

Step 6: Add Dependencies

Ensure pyproject.toml includes required packages:
pyproject.toml

Step 7: Add Crewship Configuration

Create crewship.toml in the project root:
crewship.toml

Step 8: Set Environment Variables

Step 9: Deploy

Step 10: Run Your Crew

Watch the execution:

Step 11: Get the Artifact

Next Steps

Streaming

Add real-time event handling

Artifacts

Work with run outputs

Your First LangGraph Agent

Build an agent with LangGraph (Python)

Your First LangGraph.js Agent

Build an agent with LangGraph.js