Workflow automation · 6 tools
Top 6 workflow automation tools for teams that need open-source control over their data
If your data has to stay in your own environment, start with n8n, Activepieces and Node-RED: three tools that pair self-hosting with general-purpose workflow building. Temporal, Apache Airflow and Prefect follow for teams that prefer to write orchestration as code. This shortlist is for engineering and IT teams with data sovereignty requirements.
| # | Tool | Best for | Key feature | Platforms / deployment | Pricing | More |
|---|---|---|---|---|---|---|
| 1 | n8nn8n GmbH | Technical teams that want a visual builder with code flexibility, deployed in their own environment | 500+ native integrations, plus any API via HTTP and agents via MCP | On-prem, Docker, or hosted version | See vendor pricing page | Introduction |
| 2 | ActivepiecesActivepieces | Mixed teams of non-developers and engineers that need self-hosted or air-gapped automation | Flow builder with branching, retries, waitpoints, subflows, error handling and human approvals | Vendor cloud (Frankfurt), your own servers with Docker or Kubernetes, or air-gapped | See vendor pricing page | Introduction |
| 3 | Node-REDOpenJS Foundation | Event-driven data flows, from edge devices to industrial systems | Browser-based flow editor with one-click deployment | Built on Node.js; runs at the edge on low-cost hardware such as Raspberry Pi and in the cloud | See vendor pricing page | Introduction |
| 4 | TemporalTemporal Technologies | Developers building reliable, long-running business processes in code | Workflows capture state at every step and resume after failure | Self-hosted open source or managed Temporal Cloud; local dev server via CLI on macOS, Windows and Linux | See vendor pricing page | Introduction |
| 5 | Apache AirflowApache Software Foundation | Python teams scheduling and monitoring data and infrastructure pipelines | Pipelines defined in Python, enabling dynamic generation | Not stated on official pages | See vendor pricing page | Introduction |
| 6 | PrefectPrefect Technologies | Python teams orchestrating data pipelines and AI agents with workers in their own environment | Any Python function becomes a durable workflow with the @flow decorator, with no YAML or DSL | Open-source core; managed service; workers on Kubernetes, ECS, Docker or serverless platforms such as Cloud Run and Modal | See vendor pricing page | Introduction |
1 n8n
By n8n GmbH
n8n is a workflow automation platform that combines AI capabilities with business process automation for technical teams. It offers a visual builder with the option to use code.
Why it is here: It opens the list because it pairs on-prem or Docker deployment and public source code with a broad, general-purpose builder for business workflows.
- Best for
- Technical teams that want a visual builder with code flexibility, deployed in their own environment
- Key features
- 500+ native integrations, plus any API via HTTP and agents via MCP
- On-prem or Docker deployment, with source code on GitHub
- Human-in-the-loop approvals and the ability to connect any model
- Re-run a failed step on its own; replay past runs
- Audit logs, log streaming to a SIEM and Git-based control
- Watch out for
- Enterprise features such as SAML SSO, LDAP, encrypted secrets and RBAC are listed as enterprise features, so teams that need them should confirm availability with the vendor. A hosted version also exists, so self-hosting is a deployment choice to make deliberately.
- Platforms / deployment
- On-prem, Docker, or hosted version
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
2 Activepieces
By Activepieces
Activepieces is an open-source workspace that brings chat, AI agents, flows, tables and 770+ apps together so teams can automate work with AI.
Why it is here: It sits second because it documents the widest range of deployment options for data control, from its own cloud to your servers to fully air-gapped, alongside a visual flow builder.
- Best for
- Mixed teams of non-developers and engineers that need self-hosted or air-gapped automation
- Key features
- Flow builder with branching, retries, waitpoints, subflows, error handling and human approvals
- Inline TypeScript within otherwise visual flows
- Every piece is MIT-licensed
- Self-hosting with Docker or Kubernetes, or fully air-gapped
- Credentials in external secret managers; audit trail can stream to a SIEM
- Watch out for
- The vendor describes the air-gapped option as running your own model on your own hardware, so teams should plan for that infrastructure. Compliance statements such as the SOC 2 Type 2 report are vendor-stated, and teams should verify them directly.
- Platforms / deployment
- Vendor cloud (Frankfurt), your own servers with Docker or Kubernetes, or air-gapped
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
Official website (activepieces.com) · Read our introduction to Activepieces
3 Node-RED
By OpenJS Foundation
Node-RED is a low-code, flow-based programming tool for building event-driven applications that collect, transform and visualize data. It was originally developed by IBM and is now part of the OpenJS Foundation.
Why it is here: It comes third because it is open source and runs wherever you host it, but it is built around event-driven data flows rather than business-process features such as approvals.
- Best for
- Event-driven data flows, from edge devices to industrial systems
- Key features
- Browser-based flow editor with one-click deployment
- Over 5,000 community-built nodes and flows
- JavaScript functions inside the editor
- Flows stored as JSON for import and export
- Runs on low-cost hardware like Raspberry Pi and in the cloud
- Watch out for
- The official pages describe flow editing, community nodes and sharing. Teams that need SSO, audit logging or role-based access should check the vendor documentation.
- Platforms / deployment
- Built on Node.js; runs at the edge on low-cost hardware such as Raspberry Pi and in the cloud
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
Official website (nodered.org) · Read our introduction to Node-RED
4 Temporal
By Temporal Technologies
Temporal is an open-source durable execution platform for writing business logic as code in Workflows that automatically recover from failure.
Why it is here: It begins the code-first group: it is MIT-licensed and self-hostable, and it suits teams that want business logic written in code rather than assembled visually.
- Best for
- Developers building reliable, long-running business processes in code
- Key features
- Workflows capture state at every step and resume after failure
- Activities retry automatically; retry policies are configured rather than coded
- Timers can wait days, weeks or months and survive restarts
- SDKs for Go, Java, TypeScript, Python and .NET
- Client-side encryption through the Data Converter
- Watch out for
- Temporal is documented as a code-based platform for developers, so teams without engineering capacity for it should check the vendor documentation. Self-hosting means running the Temporal Service yourself; the managed Temporal Cloud is the alternative.
- Platforms / deployment
- Self-hosted open source or managed Temporal Cloud; local dev server via CLI on macOS, Windows and Linux
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
Official website (temporal.io) · Read our introduction to Temporal
5 Apache Airflow
By Apache Software Foundation
Apache Airflow is a community-created platform to programmatically author, schedule and monitor workflows, defined in Python.
Why it is here: It follows Temporal because it is a code-first orchestrator aimed at Python users, with a focus on pipelines rather than general business workflows.
- Best for
- Python teams scheduling and monitoring data and infrastructure pipelines
- Key features
- Pipelines defined in Python, enabling dynamic generation
- Custom operators and extendable libraries
- Web application for monitoring, scheduling and managing workflows
- Plug-and-play operators for major cloud providers and third-party services
- Provider package for LLMs and AI agents
- Watch out for
- The vendor states the audience as people with Python knowledge. Deployment options and access-control features are not described on the pages we reviewed, so teams with specific requirements should check the vendor documentation.
- Platforms / deployment
- Not stated on official pages
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
Official website (airflow.apache.org) · Read our introduction to Apache Airflow
6 Prefect
By Prefect Technologies
Prefect is a Python workflow orchestration product, built on the open-source Prefect framework, for data pipelines and AI agents.
Why it is here: It closes the list because it is also code-first and Python-centred, with documented hybrid execution in which workers run in your own environment.
- Best for
- Python teams orchestrating data pipelines and AI agents with workers in their own environment
- Key features
- Any Python function becomes a durable workflow with the @flow decorator, with no YAML or DSL
- Workers poll over outbound-only connections, so no inbound network access is needed
- Pause for human input in the UI, such as approvals or forms
- Run on Kubernetes, ECS, Docker or serverless
- RBAC, SSO, SCIM, audit logs and IP allowlisting
- Watch out for
- Managed and push execution are not part of the open-source Prefect server, so teams self-hosting the open-source core should check which features they need. Several access-control and compliance features are described for the managed product.
- Platforms / deployment
- Open-source core; managed service; workers on Kubernetes, ECS, Docker or serverless platforms such as Cloud Run and Modal
- Pricing
- See vendor pricing page (not quoted here; facts checked 21 November 2026, Asia/Dhaka)
Official website (prefect.io) · Read our introduction to Prefect
How we chose these tools
We looked at three things: fit with the use case (open-source or self-hostable licensing combined with business workflow building), documented capabilities such as integrations, approvals, error handling and access controls, and the deployment and platform options each vendor documents. Tools that offer a visual builder alongside self-hosting come first, and code-first orchestration engines come last. The order reflects editorial fit for this specific use case, and we explain the reasoning in each entry. It is not a statement that one tool is better than another overall.
All facts come from official vendor pages, and we record the date each page was checked. The list is built from documentation rather than product trials, so it describes what vendors state, not how a tool behaves in your environment. Where a vendor page does not state something, we say so instead of filling the gap. Check current licensing, deployment and security details with the vendor before you commit.
Pricing is not covered here. Each tool's page shows a link to the vendor's own pricing information.
Order used in this guide: Ordered by how well a tool combines open-source or self-hostable licensing with general-purpose business workflow building, with code-first orchestration engines last.
Frequently asked questions
Which open-source workflow automation tools can be self-hosted?
Among the tools listed, n8n can be deployed on-prem or with Docker, and Activepieces supports Docker or Kubernetes. Temporal can be self-hosted as open source. Prefect's open-source core is self-hostable, with some managed features excluded from the open-source server.
Can a workflow automation tool run fully air-gapped?
Activepieces documents a fully air-gapped deployment that keeps everything inside your network and runs your own model on your own hardware. Other tools here do not describe an air-gapped option on the pages we reviewed, so check their documentation if you need one.
Which tools suit teams that prefer a visual builder over code?
n8n and Activepieces both offer visual builders with the option to add code, and Node-RED uses a browser-based flow editor. Temporal, Apache Airflow and Prefect are code-first and expect you to write workflows in a programming language.
Do these tools support human approvals in workflows?
n8n, Activepieces and Temporal all document human-in-the-loop or human-interaction support, and Prefect lets workflows pause for human input in its UI. The Airflow and Node-RED pages we reviewed do not describe approvals.
How does Prefect keep customer data in the customer's environment?
Prefect states that Prefect Cloud coordinates workflows without accessing customer code or data. Its workers poll over outbound-only connections, so no inbound network access is needed.
- n8n: n8n.io — checked 21 November 2026
- Activepieces: activepieces.com, activepieces.com/security, activepieces.com/product/chat — checked 21 November 2026
- Node-RED: nodered.org, nodered.org/about, nodered.org/about/conduct — checked 21 November 2026
- Temporal: temporal.io, temporal.io/about, temporal.io/product — checked 21 November 2026
- Apache Airflow: airflow.apache.org — checked 21 November 2026
- Prefect: prefect.io, prefect.io/security, prefect.io/how-it-works — checked 21 November 2026
Logicwick is independent and not affiliated with the vendors listed. The editorial order is unpaid. A vendor may buy one clearly labeled featured slot per guide, sold to only one company, and we still check its facts; see our disclosure. Product names belong to their owners. Features and prices change; confirm details on each official website.