agentify-toolkit 0.2.1

Declarative AI toolkit

Agent Building Toolkit - Agentify 🤖

Agentify is a lightweight, declarative-first Library for building AI agents in Python

KeyFeatures

• Declarative agent definitions via YAML.
• Multi-LLM support: OpenAI, Anthropic, and more.
• Interactive CLI with TUI menu for exploring agents.
• Programmatic agent creation and execution for custom workflows.
• Lightweight, minimal dependencies: Click + Rich + PyYAML.

🚀 Getting Started

Prefer a hands-on walkthrough?

• 📘 Agentify Developer Quickstart (Jupyter)
• 📄 Agentify YAML Deep Dive

How it works

Define your agents in simple YAML files or programmatically, and run them using an interactive CLI. Agentify abstracts LLM provider integrations and provides a simplified developer experience.

Installation

pip install agentify-toolkit

Or install from source:

git clone https://github.com/backplane-software/agentify.git
cd agentify
pip install .

Quick Start

Note: To use Agentify you will require an API KEY from your AI Model Provider.

To add the API KEY to your environment:

agentify providers add <provider> # e.g. anthropic, xai, google, openai

Then, you can validate:

agentify providers list

1. Create an agent YAML

name: claude
description: AI Engineer
version: 0.1.0
model:
  provider: anthropic
  id: claude-sonnet-4-5
  api_key_env: ANTHROPIC_API_KEY
role: |
  You are an AI Security Engineer.
  Provide concise, practical answers with examples.

2 .Run a single agent

agentify run examples/agent.yaml

Overriding Model - you can pass a provider and model parameter to override the provider and model in the YAML file

agentify run examples/agent.yaml --provider=openai --model=gpt-5-nano

3. Run agents interactively from a folder

agentify run examples/agents

4. List agents interactively

agentify list agents

Programmatic Usage

Install agentify with pip install agentify and then import.

from agentify import Agent

def main():

    agent = Agent(
        name="Grok",
        description="X's Grok Agent",
        provider="x",
        model_id="grok-4",
        role="You are an AI Security Architect specialising in X AI Grok models"
    )

    response = agent.run("Which AI LLM is the best in 1 sentence ?")
    print(response)

if __name__ == "__main__":
    main()

CLI Reference

Command	Example
Run agent from YAML	agentify run <agent.yaml>
Run all YAML Agents in folder	agentify run <folder>
List all YAML Agents in folder	agentify list <folder>

Adding LLM API Keys

To use the LLM Model providers, you will need an API key and they will need to be made available in your environment.

Implemented providers:

Provider	Instruction
OpenAI	export OPENAI_API_KEY="your_api_key_here"
Anthropic	export ANTHROPIC_API_KEY="your_api_key_here"
Gemini	export GEMINI_API_KEY="your_api_key_here"
GROK	export XAI_API_KEY="your_api_key_here"
Bedrock	export AWS_BEARER_TOKEN_BEDROCK

For Windows: $env:<Provider>_API_KEY="YOUR_API_KEY"

License

Agentify is licensed under the Apache License, Version 2.0. See LICENSE for details.