Home
Agentic makes it easy to create AI agents - autonomous software programs that understand natural language and can use tools to do work on your behalf.
What is Agentic?
Agentic is an opinionated framework that provides:
- A lightweight agent framework in the same part of the stack as SmolAgents or PydanticAI
- A reference implementation of the agent protocol
- An agent runtime built on threads with support for Ray
- A suite of "batteries included" features to help you get running quickly
Key Features
- Simple but Powerful: Approachable and easy to use, yet flexible enough for complex agent systems
- Team Collaboration: Support for teams of cooperating agents
- Human-in-the-Loop: Built-in support for human intervention and guidance
- Tool Ecosystem: Easy definition and use of tools with a growing library of production-ready tools, as well as the ability to create your own
- Flexible Deployment: Run agents as standalone applications, API services, or integrated components
- Framework Interoperability: Work with agents built using other frameworks
Quick Install
Note: Agentic requires Python 3.12. It does not work with Python 3.13+ due to Ray compatibility issues.
# Install from PyPI
pip install agentic-framework
# Or run from source
git clone git@github.com:supercog-ai/agentic.git
cd agentic
uv pip install -e ".[all,dev]"
# Initialize your project
agentic init .
Hello World Example
from agentic.common import Agent, AgentRunner
from agentic.tools.weather_tool import WeatherTool
weather_agent = Agent(
name="Weather Agent",
welcome="I can give you some weather reports! Just tell me which city.",
instructions="You are a helpful assistant.",
tools=[WeatherTool()],
model="openai/gpt-4o-mini"
)
if __name__ == "__main__":
AgentRunner(weather_agent).repl_loop()
Ready-to-Run Agents
Agentic comes with several pre-built agents you can run immediately:
- OSS Deep Researcher: Writes detailed research papers on any topic with references
- OSS Operator: Full browser automation with authenticated sessions
- Podcast Producer: Auto-produces and publishes daily podcasts
- Meeting Notetaker: Records and summarizes meetings
- Database Agent: Text-to-SQL for data analysis using natural language
Next Steps
- Getting Started Guide - Step-by-step introduction to Agentic
- Core Concepts - Understand the fundamental ideas behind the framework
- Examples - Explore real-world agent examples
- CLI Reference - Command-line interface documentation