Self-hosted platform for discovering and running AI agents

Deploy specialized agents on your infrastructure—without building from scratch or using SaaS. AgentSystems combines a federated agent ecosystem, provider portability, and container isolation to create portable AI agents that run in your environment—laptop, server, or data center.

Open Source Apache-2.0 license
# Install on macOS or Linux - follow the setup prompts
curl -fsSL https://github.com/agentsystems/agentsystems/releases/latest/download/install.sh | sh

Federated Agent Ecosystem

Git-based agent index using GitHub forks. No central gatekeepers—anyone can operate their own agent index alongside community indexes.

Provider Portability

Agents integrate with OpenAI, Anthropic, Bedrock, Ollama—write once, run anywhere. This is the "write once, run anywhere" moment for AI agents.

Container Isolation & Network Controls

Each agent runs in its own Docker container with configurable network egress filtering, thread-scoped artifact storage, and hash-chained audit logs.

Overview

See AgentSystems in Action

Watch the 100-second overview below or dive into the full 9-minute demo and walkthrough.

Why AgentSystems

Teams want to use specialized AI agents but face a dilemma

Teams need specialized AI agents—codebase migration, research synthesis, visual content analysis, structured data extraction—but face three bad options:

🔒 SaaS Agents

Require sending data to third parties

🛠️ Building from Scratch

Takes weeks of development per agent (most teams lack ML expertise)

🐳 Manual Docker Orchestration

Requires configuring networks, volumes, proxies, and API keys for each agent

AgentSystems provides a standardized runtime and ecosystem

For Teams Without Infrastructure Expertise

Single command deployment handles networking, separation, and audit logging

For Organizations Requiring On-Premises

Run agents on-premises or air-gapped with configurable egress controls

For Developers Building Products

Browse community agents instead of building from scratch

For Enterprises

Discover, evaluate, and deploy agents with container isolation and audit logging

Key Benefits

Built for Self-Hosting

Federated Agent Ecosystem

Git-based agent index where developers publish via GitHub forks. Anyone can run their own index alongside the community index. No central authority controls listing or distribution.

Container Isolation & Network Controls

Each agent runs in its own Docker container with configurable egress filtering, thread-scoped artifact storage, hash-chained audit logs, and lazy startup.

Provider Portability

Switch from OpenAI to Anthropic to Ollama through configuration. Run the same agent with different models and providers. Reduce vendor lock-in at the agent level.

Flexible Deployment

Laptop, server, Kubernetes—works where Docker runs.

Observable

Langfuse integration for tracing LLM calls and agent execution.

Audit Trail

SHA-256 hash-chained logs for operation tracking.

Thread Isolation

Each request gets a unique thread ID with separate artifacts.

Use Cases

Example Use Cases

Concrete examples of what you can build with AgentSystems

Personal Projects

Research Synthesis

Input: 20 PDFs
Output: Structured literature review with citations and theme analysis

Codebase Migration

Input: Legacy Python 2.7 codebase
Output: Python 3.12 implementation with migration notes

Content Extraction

Input: Product pages from 10 websites
Output: Structured JSON with pricing, specs, availability

Startups & Small Teams

Report Generation

Input: CSV exports + template
Output: Formatted PDF with charts, tables, executive summary

Data Normalization

Input: Bank statements from 5 institutions
Output: Standardized transaction records in unified format

Documentation Generation

Input: Codebase repository
Output: API reference docs with examples and descriptions

Enterprise & On-Premises

Contract Analysis

Input: 100-page legal documents
Output: Structured data extraction with clause identification

Security Audit Workflows

Input: Application codebase
Output: Vulnerability report with findings

Report Generation

Input: Internal databases + requirements
Output: Structured reports

Browse available agents at agentsystems.github.io/agent-index

Quick Start

Deploy in Minutes

Get AgentSystems running with a single command

# One-command install (macOS/Linux)
curl -fsSL https://github.com/agentsystems/agentsystems/releases/latest/download/install.sh | sh

# Initialize and start
agentsystems init agentsystems-platform && cd agentsystems-platform
agentsystems up

# Open web UI
open http://localhost:3001

What Gets Installed

  • agentsystems CLI
  • Docker (if not present)
  • Required dependencies

What Starts

  • Gateway (port 18080)
  • Web UI (port 3001)
  • PostgreSQL
  • Agent containers

Requirements

  • macOS or Linux
  • Works on laptops, servers, VMs
  • Air-gapped capable
Comparison

How AgentSystems Compares

vs. LangChain/CrewAI

AgentSystems is a runtime for agents built with these frameworks. Your agent code uses LangChain; AgentSystems deploys and runs it.

vs. Portkey/LiteLLM

Different layers. AI gateways route API calls; AgentSystems runs complete applications with workflows and file I/O.

vs. Manual Docker

You can build this yourself. AgentSystems is the pre-built, standardized version with discovery, egress control, audit logs, and artifact management.

For Agent Developers

Build Once, Reach Self-Hosting Customers

Some customers won't use your agent if it means sending data to your servers. AgentSystems gives you a distribution channel for those customers.

Build Your Agent

Use standard frameworks like LangChain and LangGraph. Wrap in a FastAPI app with /invoke, /health, and /metadata endpoints.

Publish to the Index

Git-based publishing via GitHub forks. No approval process—automated validation only.

Customers Deploy on Their Infrastructure

Your agent runs on their infrastructure under their control. You don't host or store their data. You focus on agent capabilities, they handle deployment.

Note: AgentSystems does not review or endorse software in any index. All agents are published by third-party developers.
Components

Integrated Platform Components

Core components work together as a complete system

Control Plane

FastAPI gateway managing agent lifecycle, routing, and audit logs.

View Repository

CLI & SDK

Command-line tool for deployment and management.

View Repository

Web UI

React dashboard for agent management and monitoring.

View Repository

Python Toolkit

Library for building agents with model routing.

View Repository

Agent Template

FastAPI starter with tracing and artifact handling.

View Repository

Agent Index

Federated, Git-based registry for agent discovery.

View Repository

Documentation

Guides for setup and development.

View Docs
Community

Join the Open Agent Ecosystem

Build, share, and discover AI agents with a growing community of developers.

Contributors

Contribute to the development of self-hosted AI infrastructure.

Contributing Guide

Publish Agents

List your agents on the public index for community discovery.

Publishing Guide

Discord Community

Get help, share agents, and shape the roadmap with other users.

Join Discord

Report Issues

Found a bug or have a feature request? Let us know on GitHub.

Open Issue

Contact Us

Questions about AgentSystems? Need help with deployment?

Get in Touch