heymrun/heym

Overview

Heym is a self-hosted AI workflow automation platform built in Python. It provides a visual canvas for building agents, RAG pipelines, human-in-the-loop steps, and MCP integrations, all within a single runtime with built-in observability.

Best for

Best for
Developers who need a self-hosted, all-in-one platform for building and monitoring complex AI workflows.

Use cases

• Orchestrating multi-step AI workflows with a visual drag-and-drop canvas
• Building retrieval-augmented generation pipelines with human review steps
• Deploying self-hosted agent systems with monitoring and debugging

How to use

Install

./run.sh

Tools exposed

• DATABASE_URL
• POSTGRES_HOST
• POSTGRES_PORT
• SECRET_KEY
• BACKEND_PORT
• FRONTEND_PORT
• ALLOW_REGISTER
• REQUEST_BODY_MAX_SIZE_MB
• HEYM_OTEL_ENABLED
• HEYM_OTEL_EXPORTER_OTLP_ENDPOINT
• HEYM_OTEL_EXPORTER_OTLP_HEADERS
• HEYM_OTEL_SERVICE_NAME
• HEYM_OTEL_TRACES_SAMPLER_RATIO
• HEYM_OTEL_CAPTURE_NODE_IO
• HEYM_MCP_ALLOW_PRIVATE_URLS

Tested with

Claude Desktop, Cursor, Continue, ChatGPT

Notes

Heym is a self-hosted AI workflow automation platform built in Python. It provides a visual canvas for building agents, RAG pipelines, human-in-the-loop steps, and MCP integrations, all within a single runtime with built-in observability.

565 stars on GitHub. Last updated 2026-06-01.

Use cases

• Orchestrating multi-step AI workflows with a visual drag-and-drop canvas
• Building retrieval-augmented generation pipelines with human review steps
• Deploying self-hosted agent systems with monitoring and debugging

Pros

• Self-hosted, giving full control over data and infrastructure
• Combines agents, RAG, HITL, and MCP in one runtime
• Includes observability for debugging and monitoring workflows

Cons

• Relatively new project with a small community (565 stars)
• Requires self-hosting setup and maintenance
• Limited documentation and ecosystem compared to established platforms

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.