OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Multi-Agent Systems

Agentswarms

Agentswarms is a self-hosted AI and data platform for building AI agents and multi-agent swarms alongside data pipelines, a lakehouse, dashboards, an AI analyst and no-code machine learning. It combines workflow design and data analysis in one application backed by Supabase.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agentswarms — OpenViST project profile in Multi-Agent Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a team wants to build agent workflows and connect them to its own data platform, and is prepared to deploy and operate the stack itself.

Who it is for

  • Teams building and operating multi-agent workflows on infrastructure they control.
  • Data teams connecting ETL pipelines and lakehouse data to dashboards or conversational analysis.
  • Operators seeking a self-managed environment for agent and analytics tools.

Documented uses

  • Design multi-agent workflows as visual graphs with agent, router, condition, loop, approval and tool-call steps.
  • Create RAG knowledge bases and use an AI analyst to ask questions over platform data, inspect generated SQL and review cited findings.
  • Build ETL pipelines visually or in Python to move and transform data into destinations.
  • Explore lakehouse tables and snapshots, query data, and build BI dashboards or no-code machine-learning workflows.
  • Deploy the application with Docker or Kubernetes and configure its Supabase backend.
Before you choose

Requirements & limits

The repository documents self-hosted deployment using Docker or Kubernetes. Setup can use an existing Supabase project or a self-hosted Supabase stack.

  • The documented stack includes Node.js 20.19 or later, React 19, TanStack Start and Supabase; deployment options include Docker and Kubernetes.
  • The self-hosted Supabase setup documents roughly 2 GB of image pulls and an additional 2 vCPU and 4 GB RAM for the Supabase stack.
  • Agent workflow drafts and published snapshots are distinct: API keys and schedules serve the last published snapshot until it is promoted.
  • The project describes model-provider and data-source integrations, but model keys and infrastructure must be configured by the operator.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Agentswarms with self-hosted Supabase

Prepare the application and Supabase stack on a Docker-capable environment using the documented setup script.

Before you start

  • Docker running; on Windows, use WSL or Git Bash with Docker Desktop.
  • Allow about 2 GB of image pulls and an additional 2 vCPU and 4 GB RAM for the Supabase stack.

Steps

  1. Run the command below from the project repository.

    Documented command
    bash scripts/setup-selfhosted.sh

  2. Wait for the script to start Supabase, generate secrets and keys, apply the schema, create an admin user and bring up the application.

  3. Open the local application at http://localhost:8080. The documented non-interactive option sets ADMIN_EMAIL before running the script.

Expected outcome

A local application and self-hosted Supabase stack, with an admin user created by setup.

Guide limitations

  • The guide documents setup, not the configuration or execution of a particular agent workflow.
  • The Windows instructions require WSL or Git Bash and Docker Desktop.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
AgentSwarms-fyi/agentswarms · Official homepage title · retained in its source language
How we review sourcesSuggest a correction