Best open-source alternatives to Make (Integromat)

Huginn is a self-hosted automation platform for people who want to build their own event-driven agents instead of relying on hosted services like IFTTT or Zapier. It lets users create agents that read from the web, watch for changes or signals, and trigger downstream actions, all while keeping the data on their own server. The project is aimed at technical users who want flexible, hackable automations for tasks such as weather alerts, social media monitoring, web scraping, notifications, and custom workflows. It also supports extending the system with external agent gems, and the README points to Docker-based and manual/server-based installation paths for getting started and deployment.

Langflow is an open source platform for creating AI-powered agents and workflows. It is aimed at developers who want a visual way to assemble flows while still being able to customize components in Python and integrate the result into applications. The project supports interactive testing, multi-agent orchestration, deployment as an API, and conversion of flows into MCP servers. It also mentions support for major LLMs, vector databases, and observability integrations, positioning it as a flexible tool for building and operationalizing AI applications across different stacks.

Node-RED is a low-code development tool for building event-driven applications. It is aimed at developers and makers who want to wire together data sources, services, and logic visually rather than writing everything from scratch. The project is run as a Node.js application and can be started locally after installing it with npm. Its ecosystem includes a library of nodes, reusable flows, and collections, along with documentation and community support through the forum and Slack. The README also points to custom nodes and integrations, suggesting that the platform is designed to be extended and adapted to many automation and integration use cases.

Kestra is an open-source orchestration platform for automating workflows that are either scheduled or triggered by events. It is designed for teams building data pipelines, process automation, and microservice orchestration, with a strong emphasis on Infrastructure as Code and declarative definitions. The project centers on YAML-based flows that can be created and edited from a web UI, then synced back to version control or managed via API and infrastructure tooling. Its plugin ecosystem lets users run scripts, query databases, interact with cloud services, and respond to events from many external systems, while the UI provides visualization and editing assistance for building and operating workflows.

Trigger.dev is an open-source platform for building AI workflows and agents in TypeScript. It is aimed at developers who want to create background jobs and long-running tasks that need retries, queues, observability, and elastic scaling without managing infrastructure themselves. The project provides an SDK for defining tasks directly in a codebase, along with features such as durable execution, realtime updates, human-in-the-loop pauses, and monitoring. It can be used through Trigger.dev's cloud service, or self-hosted using Docker Compose or Kubernetes with the official Helm chart. The README also highlights environments for development, preview, staging, and production, making it suitable for teams shipping workflow-driven applications.

Sim is an open-source platform for creating AI agents and operating agentic workflows. It is aimed at users who want to visually design, connect, and run automation flows that combine agents, tools, integrations, and language models. The project supports a cloud-hosted experience as well as self-hosted deployment options. In self-hosted setups, it can be run through Docker Compose, an NPM package, or manual installation, and it also supports local AI models through Ollama and vLLM. The README also describes Copilot features for generating nodes and helping users iterate on workflows, along with knowledge upload and retrieval for grounding agent responses in user-provided documents.

StackStorm is an automation platform designed for connecting services and tools and reacting to events with predefined actions. It is aimed at operations and infrastructure teams that want to automate troubleshooting, remediation, and deployment-style workflows across systems such as monitoring tools, cloud platforms, chat systems, and ticketing services. The project organizes automation into sensors, triggers, actions, rules, workflows, and packs. Content is managed as code, which makes it suitable for collaboration and sharing through the StackStorm Exchange. StackStorm can be operated through a REST API, CLI client, and web UI, and its modular architecture is built from microservice components communicating over a message bus.

Budibase is an open-source operations platform designed to help teams build AI-driven agents, apps, and automations. It is aimed at organizations that want to reduce manual operational work by connecting requests, approvals, processes, and business systems in one self-hostable platform. The project supports both cloud-hosted and self-hosted deployments, with documentation for Docker, Docker Compose, Kubernetes, and Digital Ocean. Its public API is positioned for extensibility and interoperability, and the repository also points to modular server, builder, and client packages that power the overall system. Budibase appears to be built for collaborative, multi-user operational environments. The README highlights global management of users, onboarding, SMTP, apps, groups, theming, and delegation of user management to group managers, while also noting paid features under a Business Source License.

Activepieces is an open source automation platform designed as an alternative to Zapier. It combines a visual, no-code workflow builder with an extensible pieces framework written in TypeScript, allowing developers to create integrations and logic that can be reused across the platform. The README emphasizes both technical flexibility and accessibility for non-technical users. The project is aimed at teams and organizations that want to automate workflows, connect services, and customize their automation stack while keeping control over deployment. It supports a broad integration ecosystem, AI-oriented pieces, human-in-the-loop interactions, and MCP availability for use with LLM tools such as Claude Desktop, Cursor, and Windsurf. The community edition is MIT licensed, while enterprise features are offered under a separate commercial license.

ToolJet is an open-source foundation for ToolJet AI, positioned as a platform for building and deploying internal tools, workflows, and AI agents. The README presents the community edition as a visual, low-code builder aimed at teams that need to assemble internal applications quickly, connect them to business data, and extend them with custom logic. It appears to target organizations that want self-hosted internal tooling with collaboration and security features. The community edition emphasizes a drag-and-drop editor, a built-in database, broad connector support, and the ability to run JavaScript and Python inside apps. The enterprise-facing ToolJet AI layer adds AI-assisted app creation, query building, debugging, role-based access control, advanced permissions, multi-environment management, Git sync, and white-label customization.