Activepieces
Open-source no-code automation platform with integrations, AI steps, and MCP servers.
MIT core, separately licensed enterprise features self-hosted or vendor cloud
AI Agent Platforms / Open-source and self-hostable workflow automation platforms
Open-source and self-hostable workflow automation platforms
Workflow automation platforms run repeatable processes from triggers, schedules, and integrations, and many now include AI steps or embedded agents. This page lists the open-source and source-available options, with the license and hosting model for each. Read license terms carefully here, because several use fair-code or source-available licenses rather than standard open source. There are 14 in this directory.
What to look for
- Breadth of prebuilt integrations with the services you use
- AI steps and the ability to embed agents inside a workflow
- Code steps for logic the visual nodes cannot express
- Self-hosting support and how much is gated behind an enterprise license
- Triggers and scheduling for event-driven and recurring runs
- The exact license, since fair-code terms restrict some commercial uses
Platforms
Apache Airflow
Platform to programmatically author, schedule, and monitor workflows defined as Python code.
Apache-2.0 self-hosted; third-party managed services available
Automatisch
Business automation tool that connects services like Twitter and Slack to run workflows.
AGPL-3.0 core, separately licensed enterprise files self-hosted; vendor cloud available
Dagster
Python orchestrator that models data pipelines as declarative software-defined assets.
Apache-2.0 self-hosted; vendor cloud available
Huginn
Self-hosted system of agents that read the web, watch for events, and act on your behalf.
MIT self-hosted
Kestra
Event-driven orchestration platform with YAML workflows for data, AI, and infrastructure.
Apache-2.0 core, separately licensed enterprise features self-hosted or vendor cloud
n8n
Fair-code platform for building AI agents and workflows on a visual canvas with code.
Sustainable Use License (fair-code, source-available) self-hosted or vendor cloud
Node-RED
Low-code, flow-based programming tool for wiring devices, APIs, and online services.
Apache-2.0 self-hosted
Pipedream
Hosted integration platform for event-driven automations with prebuilt and custom code steps.
Pipedream Source Available License vendor-hosted
Prefect
Python workflow orchestration framework for building resilient data pipelines.
Apache-2.0 self-hosted server or vendor cloud
StackStorm
Event-driven automation platform that maps triggers to actions through rules and workflows.
Apache-2.0 self-hosted
Temporal
Durable execution platform that runs workflows resiliently and retries failed operations.
MIT self-hosted or vendor cloud
Trigger.dev
Open-source TypeScript platform for background jobs and long-running AI workflows.
Apache-2.0 self-hosted or vendor cloud
Windmill
Developer platform that turns scripts into APIs, background jobs, workflows, and UIs.
AGPL-3.0 and Apache-2.0 core, separately licensed enterprise features self-hosted or vendor cloud
Common questions
How do workflow automation platforms differ from AI agents?
Automation platforms execute flows you define in advance, which makes them predictable. Agents decide their own steps toward a goal. Many teams combine both: a workflow handles the reliable parts and calls an agent for the open-ended step.
Are these platforms fully open source?
Some are, and some are source-available or fair-code, which publishes the code but restricts certain uses. Each entry states its exact license so you can judge fit before adopting it.
Other categories
Open-source AI coworkers and teammatesOpen-source agent builders and frameworksOpen-source browser agentsOpen-source coding agents
See also the guide to choosing a platform and the full landscape and market map.