ABZ AgentABZ AgentDocs← Site

Welcome to ABZ Agent SDK

Welcome to the official documentation for ABZ Agent SDK.

ABZ Agent SDK is a lightweight, developer-first Python framework for building intelligent AI agents with a clean and intuitive API.

Whether you're creating a simple chatbot, an automation workflow, a research assistant, or a production-ready multi-agent system, ABZ Agent SDK provides the tools you need while staying simple to learn and enjoyable to use.

Our goal is straightforward:

Help developers spend less time fighting frameworks and more time building AI applications.

Why ABZ Agent SDK?#

Many AI frameworks become difficult to understand as projects grow.

ABZ Agent SDK focuses on keeping the developer experience simple without sacrificing flexibility.

With just a few lines of Python, you can build an AI agent that can reason, use tools, remember conversations, generate structured data, and integrate with modern language models.

Features#

Multi-Provider Support
One consistent API for Google Gemini and Groq — the provider is auto-detected from the model name.
Simple Agent API
Define an agent with a name, instructions, and a model — start running it in a few lines.
Built-in Memory
A conversation buffer that remembers every turn and replays it as context automatically.
Powerful Tools
Turn any Python function into a tool with @function_tool, or build one from the Tool base class.
Multi-Step Reasoning
Let an agent call tools and reason across multiple iterations with max_iterations.
Structured Output
Parse responses directly into a Pydantic model with output_type.
Guardrails
Validate input and output before it ever reaches the user.
Agent Handoffs
Let one agent transfer a conversation to a specialist agent.
Agent as a Tool
Wrap any agent as a tool so another agent can call it.
CLI Tooling
Scaffold a new project and run agent files straight from the terminal.

What You Can Build#

AI Chatbots
Customer Support Agents
Research Assistants
Coding Assistants
Email Automation
Document Q&A
Voice Assistants
Workflow Automation
Multi-Agent Systems
Personal AI Assistants

Documentation Guide#

We recommend reading the documentation in this order:

  1. Installation
  2. Quick Start
  3. Agents
  4. Tools
  5. Memory
  6. Structured Output
  7. Guardrails
  8. Handoffs

Next Step#

Continue to the Installation page.

Next →
Installation