Quickstart

Install 💻

$ pip install crowe-b

Crowe Chat

$ crowe /

# Say hello
crowe /hi

# Ask a question
crowe /how is the weather today?

Using uv (Recommended)

uv is a fast Python package installer and resolver, written in Rust.

# Install uv
$ curl -LsSf https://astral.sh/uv/install.sh | sh

# Install crowe using uv
$ uv pip install crowe

Using poetry

# Install poetry if you haven't already
$ curl -sSL https://install.python-poetry.org | python3 -

# Add crowe to your project
$ poetry add crowe

---

Environment Configuration

OPENAI_API_KEY=""
WORKSPACE_DIR="agent_workspace"
ANTHROPIC_API_KEY=""
GROQ_API_KEY=""

🤖 Your First Agent¶

An Agent is the fundamental building block of a croweo—an autonomous entity powered by an LLM + Tools + Memory.

from crowe import Agent

# Initialize a new agent
agent = Agent(
    model_name="gpt-4o-mini", # Specify the LLM
    max_loops=1,              # Set the number of interactions
    interactive=True,         # Enable interactive mode for real-time feedback
)

# Run the agent with a task
agent.run("What are the primary advantages of adopting a multi-agent system?")

🤝 Your First Crowe: Multi-Agent Collaboration A Crowe consists of multiple agents working together. This simple example creates a two-agent workflow for researching and writing a blog post.

from crowe import Agent, SequentialWorkflow

# Agent 1: The Market Analyst
analyst = Agent(
    agent_name="Market Analyst",
    system_prompt="Analyze the target audience and recent market trends, then produce a concise insights report.",
    model_name="gpt-5",
)

# Agent 2: The Content Strategist
strategist = Agent(
    agent_name="Content Strategist",
    system_prompt="Using the insights report, craft a persuasive and creative product launch announcement.",
    model_name="gpt-5",
)

# Create a sequential workflow where the analyst's output becomes the strategist's input
workflow = SequentialWorkflow(agents=[analyst, strategist])

# Execute the workflow for a given task
launch_announcement = workflow.run("Launch of an eco-friendly smart home assistant")
print(launch_announcement)