Analysis updated 2026-05-18
Build a multi step AI agent workflow visually without writing glue code.
Connect a workflow to OpenAI, Gemini, or DeepSeek with automatic fallback between providers.
Give an agent access to your own documents through a knowledge base node for retrieval based answers.
Self host the whole platform on your own server with one install script.
| nmamizerov/assemblix | 0c33/agentic-ai | adennng/stock_strategy_lab | |
|---|---|---|---|
| Stars | 14 | 14 | 14 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | hard | hard |
| Complexity | 3/5 | 4/5 | 4/5 |
| Audience | vibe coder | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Needs Docker with Compose v2, the install script generates required secret keys automatically.
Assemblix is a visual platform for building AI agents and workflow automations. Instead of writing code, you drag nodes onto a canvas and wire them together to create a working AI workflow, for example a chain that starts, calls an agent, checks a condition, makes an HTTP request, and ends. You can then run that workflow and watch it execute in real time. Workflows are built as node graphs on a canvas powered by React Flow. Agent nodes can call one of several AI model providers, including OpenAI, Google Gemini, and DeepSeek, with automatic retries and fallback if one provider fails. Condition nodes let you branch logic, HTTP request nodes let an agent call outside APIs safely, and knowledge base nodes let agents pull in your own documents for retrieval augmented answers. The platform also supports multiple organizations and projects, stores chat sessions, and includes monitoring features like health checks and per step cost tracking for the AI calls a workflow makes. Assemblix is source available under an MIT license with a Commons Clause, meaning it is free to use, modify, and run yourself, but a separate paid Enterprise license covers billing and payment features, which are turned off by default. Getting started takes about a minute if you have Docker installed. A single install script clones the project, generates the required secret keys, writes the configuration file, and starts the whole stack for you, ending with a working app at localhost port 8080 where you create an account and build your first workflow. If you would rather set things up by hand, the README also documents a manual Docker Compose setup, a development mode with live reload for the web and API pieces, and a way to run everything natively on your machine using Postgres in Docker alongside locally run frontend and backend processes. An optional Redis and worker queue tier can be turned on later if you want workflows to run across multiple machines instead of one. The project is also built to be extended: new node types can be added as separate installable packages that get picked up automatically, without changing the core Assemblix code or needing a database migration.
A drag and drop visual builder for AI agent workflows that you can self host with a single Docker command.
Mainly Python. The stack also includes Python, React, Docker.
MIT with a Commons Clause, free to use and self host, but a separate paid license covers built in billing features.
Setup difficulty is rated moderate, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.