Heym

v1.0.0Coding Agentsstable

Self-hosted, source-available AI workflow automation platform. Build multi-agent, RAG, and tool-using pipelines on a visual canvas and publish any workflow as an MCP server (stdio/SSE/Streamable HTTP). Also an MCP client via the agent node.

aiai-agentsai-agents-frameworkai-assistantautomation
Share:
429
Stars
0
Downloads
0
Weekly
0/5

What is Heym?

Heym is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to self-hosted, source-available ai workflow automation platform. build multi-agent, rag, and tool-using pipelines on a visual canvas and publish any workflow as an mcp server (stdio/sse/streamable http)...

Self-hosted, source-available AI workflow automation platform. Build multi-agent, RAG, and tool-using pipelines on a visual canvas and publish any workflow as an MCP server (stdio/SSE/Streamable HTTP). Also an MCP client via the agent node.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Self-hosted, source-available AI workflow automation platfor

Use Cases

Visual AI workflow canvas
RAG and multi-agent pipelines
Publish as MCP server
heymrun

Maintainer

LicenseMIT License
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx heym

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Heym

Heym is a self-hosted, source-available AI workflow automation platform with a visual canvas for building multi-agent, RAG, and tool-using pipelines — and the ability to publish any workflow as an MCP server (supporting stdio, SSE, and Streamable HTTP transports). It ships with 30+ integration nodes (HTTP, Slack, Email, Redis, Playwright browser automation, Grist, webhooks, and cron), a built-in RAG system backed by QDrant, LLM cost tracking, an evals framework, and a portal feature that turns any workflow into a public chat interface. Developers use it as an open, inspectable alternative to closed no-code AI tools, particularly when they need full observability into every LLM call and tool result.

Prerequisites

  • Bun 1.0 or later
  • Python 3.11 or later
  • UV package manager for Python dependencies
  • Docker and Docker Compose
  • PostgreSQL 16 (included in the Docker Compose setup)
1

Clone the repository

Clone the Heym repository to your server or development machine.

git clone https://github.com/heymrun/heym.git
cd heym
2

Configure environment variables

Copy the example env file and set the required SECRET_KEY and ENCRYPTION_KEY values. These are used for JWT signing and data encryption respectively.

cp .env.example .env
# Edit .env and set:
# SECRET_KEY=$(python3 -c "import secrets; print(secrets.token_hex(32))")
# ENCRYPTION_KEY=$(python3 -c "import secrets; print(secrets.token_hex(32))")
3

Start the platform

Run the provided startup script, which starts the backend, frontend, and PostgreSQL database.

./run.sh
4

Open the visual canvas

Navigate to the Heym UI in your browser (default port 4017). You can build workflows by dragging nodes onto the canvas or use the AI Assistant to describe a workflow in natural language.

# Open in browser:
# http://localhost:4017
5

Publish a workflow as an MCP server

Once a workflow is built and tested, enable the MCP server toggle in the workflow settings. Heym will generate a stdio, SSE, or Streamable HTTP endpoint that external MCP clients can connect to.

6

Deploy to production with Docker

For production deployments, use the deploy script which builds and starts all services via Docker Compose.

cp .env.example .env
./deploy.sh

Heym Examples

Client configuration

After publishing a Heym workflow as an MCP server, connect to it from Claude Desktop using the generated stdio command or SSE URL. Replace the command path with the one shown in your Heym workflow MCP settings.

{
  "mcpServers": {
    "heym": {
      "command": "npx",
      "args": ["heym"]
    }
  }
}

Prompts to try

Natural language prompts for the Heym AI Assistant canvas builder and the published workflow chat portal.

- "Create a workflow that includes a Roadmap Agent and a Best Food Agent, invoked in parallel by an Orchestrator Agent."
- "Add a skill to the agent that creates a simple execution plan after the subagents return their results."
- "I live in Berlin and am planning a trip to Frankfurt. How far is it on the Autobahn? Where's the best doner?"
- "Set up a webhook trigger that runs an LLM node and emails me the result every time a form is submitted."

Troubleshooting Heym

./run.sh fails with a missing dependency error

Ensure Bun, Python 3.11+, UV, and Docker are all installed and on your PATH. Run 'bun --version', 'python3 --version', 'uv --version', and 'docker --version' to verify. Install any missing tools before re-running ./run.sh.

The database connection fails at startup

If using an external PostgreSQL instance, confirm DATABASE_URL in .env is correct. If using the bundled Docker Compose database, run 'docker compose up -d postgres' first and wait for it to be healthy before starting Heym.

Workflows published as MCP servers are not visible to external clients

Check that the FRONTEND_PORT and BACKEND_PORT values in .env are accessible from the client machine (not blocked by a firewall). For SSE transport, confirm the client is connecting to the correct URL path generated in the Heym MCP settings panel.

Frequently Asked Questions about Heym

What is Heym?

Heym is a Model Context Protocol (MCP) server that self-hosted, source-available ai workflow automation platform. build multi-agent, rag, and tool-using pipelines on a visual canvas and publish any workflow as an mcp server (stdio/sse/streamable http). also an mcp client via the agent node. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Heym?

Follow the installation instructions on the Heym GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Heym?

Heym works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Heym free to use?

Yes, Heym is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "heym": { "command": "npx", "args": ["-y", "heym"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Heym?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides