Heym

Heym is an AI workflow automation platform for visual, low-code automations with agents, RAG, MCP, and human review. Source-available and self-hosted.

AI 에이전트 개발 AI 개발 도구 노코드 로우코드 워크플로우 SOP 관리

웹사이트 방문

What Heym is

Heym is an AI workflow automation platform with a visual editor for building no-code and low-code automations. The site positions it as an AI-native runtime for workflows that combine LLM nodes, agent nodes, RAG pipelines, tool calls, and human review in one place.

Users can describe a workflow in natural language and let the assistant assemble it, or build the flow directly on the canvas from built-in nodes and integrations. Heym is source-available and fully self-hosted, so teams can run it on their own infrastructure and keep data, execution, and deployment under their control.

Core features

Visual workflow editor

Build automations on a visual canvas or start from natural-language instructions. The product also supports manual wiring from a library of nodes and integrations.

Multi-agent orchestration

Coordinate multiple agents, sub-agents, skills, and sub-workflows from one orchestrator, with human review checkpoints where needed.

Built-in RAG

Use built-in RAG pipelines for semantic search, with Qdrant support and built-in Postgres pgvector storage mentioned on the site.

Broad integration support

Connect to external tools and services through MCP client/server integration, HTTP APIs, Slack, email, GitHub, Telegram, and more.

Production workflow controls

Run workflows with tracing, evals, content guardrails, parallel DAG execution, and visible branches, tool calls, memory, and approval steps.

Source-available self-hosting

Self-host on your own infrastructure with Docker Compose or Kubernetes, and deploy with source access for customization and auditability.

Common workflows

Human-reviewed support drafting
Route support replies through an approval step so a human can review the draft before it is sent, which fits customer support teams that need guardrails.
Developer automation with GitHub
Orchestrate a coding task from plain English, use GitHub MCP to write and commit code, and post a summary to Slack for handoff or status updates.
Adversarial code review
Run review workflows that compare findings from multiple agents before producing a final approve, comment, or request-changes decision.
Content generation and publishing
Automate content pipelines that read inputs from Google Sheets, generate media with an external API, and publish the results to a destination service.
Knowledge search and RAG workflows
Build retrieval-based assistants or internal tools that search private knowledge with RAG over Qdrant or Postgres pgvector.

Pros and Cons

Pros

Supports both natural-language workflow creation and manual canvas editing.
Combines LLM, agent, RAG, MCP, and human-in-the-loop controls in one platform.
Self-hosted deployment keeps automations on infrastructure you control.
Public templates show concrete automation patterns for support, code review, and content workflows.
The pricing page states there are no licensing fees and no usage limits.

Cons

The public site does not provide a full feature-by-feature breakdown of every node type or integration.
Source availability and self-hosting add operational responsibility for deployment and maintenance.

FAQ

What is Heym used for?

Heym is a visual AI workflow automation platform. You can describe a workflow in natural language and have the assistant build it, or wire workflows manually on the canvas with built-in nodes and integrations.

How do you get started locally?

The source describes local setup as three steps: clone the repository, copy the example environment file, and run the setup script. That script starts PostgreSQL, runs database migrations, and launches the FastAPI backend and Vue.js frontend.

Who is Heym best suited for?

Heym is aimed at developers and teams that want control over their automation stack. The site emphasizes source availability, self-hosting, tracing, evals, and support for production workflows.

Can Heym be self-hosted?

The platform supports source-available self-hosting on your own infrastructure using Docker Compose or Kubernetes, and it also mentions a prebuilt container image.

Is there a paid plan or usage limit?

The pricing page says Heym is free to use with no licensing fees and no usage limits, while also noting that it is licensed under Commons Clause plus MIT, which means it is free to use, modify, and distribute but not to resell as a paid service.

Quick Facts

Category: AI workflow automation platform
Platform: Web app with self-hosted deployment options
Primary users: Developers and teams building agentic workflows
Deployment: Docker Compose, Kubernetes, or container image
Source model: Source-available
Pricing signal: Free to use; no licensing fees or usage limits stated