OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Multi-Agent Systems

Agent Kernel

Agent Kernel is a Python microkernel framework for developing large-scale multi-agent systems and adaptive social simulations powered by large language models. It provides core Agent, Environment, Action, Controller and System modules, plus plugins for scenario-specific functions. The framework is built around runtime changes: developers can add or remove LLM-based agents, alter environments and behaviours, and inter

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agent Kernel — OpenViST project profile in Multi-Agent Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a project needs coordinated LLM agents operating in a simulated social environment, especially where population size, events, environments or behaviours must change during the simulation.

Who it is for

  • Researchers studying multi-agent systems, social behaviour or collective intelligence
  • Developers building LLM-based social simulation environments
  • Teams creating reusable simulation components across multiple scenarios
  • Experimenters modelling changing populations, environments or institutional workflows

Documented uses

  • Build simulations with dynamically changing LLM-agent populations
  • Study collective behaviour and scaling effects
  • Modify simulation parameters, events or agent behaviour during runtime
  • Validate agent actions against defined physical and social rules
  • Reuse modular simulation code across different scenarios
Before you choose

Requirements & limits

The project is publicly available on GitHub under the Apache-2.0 licence. The repository lists Python 3.11 or newer as a requirement and documents Linux, Windows and macOS-oriented use.

  • Runtime addition and removal of agents is a central design capability.
  • The Controller supports intervention in parameters and events during simulation.
  • The microkernel and plugin model separates reusable core components from scenario-specific functions.
  • The framework includes system-level verification of agent actions against physical and social rules.
  • The documented examples span social, campus and clinical simulation scenarios.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Agent Kernel’s documented prerequisites and choose an installation path

Prepare the documented Python environment for using the framework.

Before you start

  • Python 3.11 or newer
  • A supported Linux, Windows or macOS environment

Steps

  1. Install or select Python 3.11 or newer.

  2. Open the Agent Kernel repository and follow its Quick Start installation section.

  3. Choose between the installation options presented in the repository documentation.

  4. If needed, follow the documented optional steps for starting Society-Panel.

Expected outcome

A Python environment prepared according to the repository’s Quick Start instructions.

Guide limitations

  • This is setup guidance, not a verified execution test.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
4 October 2026
Original website title
ZJU-LLMs/Agent-Kernel · Official homepage title · retained in its source language
How we review sourcesSuggest a correction