Heym Features

Explore Heym features: AI Assistant, multi-agent orchestration, built-in RAG, execution traces, portal chat, evaluations, and the visual workflow dashboard.

Purpose-Built for AI Workflows

Not just automation tools with AI added on top. Heym is built from the ground up with AI as the execution model.

Automatic Context Compression

Agent nodes automatically compress conversation history near the model context limit, preserving the system prompt and key turns while summarizing the middle for long-running tasks.

Human-in-the-Loop

Agent nodes can pause at approval checkpoints, generate a public review link, and wait for a reviewer to accept, edit, or refuse before resuming execution.

Guardrails

Block unsafe or unwanted content before it reaches an LLM. Choose categories like violence, hate speech, or harassment and set sensitivity to low, medium, or high per node.

MCP Support

Connect agents to any MCP server to gain external tools, or expose your Heym workflows as an MCP server for Claude Desktop, Cursor, and other clients.

Portal

Turn any workflow into a public chat UI at a custom slug URL. Supports streaming responses, file uploads, image output display, and optional per-user authentication.

Skills System

Portable capability bundles consisting of a SKILL.md instruction file and optional Python tools. Drag and drop a .zip or .md onto any Agent node, or use AI Build to draft and revise skills with live file previews.

Auto Heal

When a Playwright selector breaks at runtime, the AI step automatically finds an alternative selector and retries the action, keeping browser automations resilient to UI changes.

Visual Schedule View

See all active cron workflows on a day, week, or month calendar. Each block shows the workflow name and cron expression on hover — click to jump straight to the canvas.

Parallel Execution

The engine builds a directed acyclic graph and runs independent nodes concurrently with a thread pool. Streaming mode emits node events as each step completes, including webhook SSE consumers using custom configurable endpoints.

Enterprise Security

JWT authentication with HttpOnly cookies and refresh token rotation. Credentials encrypted at rest with AES-256 Fernet. Rate limiting on login, register, and portal endpoints.

Enterprise Integrations

Connect workflows to Telegram, Slack, IMAP, outbound WebSocket, Google Sheets, BigQuery, and more with first-party nodes for data, messaging, and realtime sync.

Error Handling & Retry

Define per-node error handlers and retry policies to recover from transient failures automatically. Workflow-level error handlers catch unhandled exceptions and route them to a dedicated recovery flow.

See Heym in Action

A powerful, intuitive interface designed for building complex AI workflows.

From the dashboard you manage workflows, credentials, vector stores, teams, and analytics. The visual canvas editor lets you connect many built-in node types with a drag-and-drop interface, pin node outputs for debugging, and use the AI Assistant to generate workflows from natural language.

The interface supports both light and dark themes, responsive layouts for desktop and tablet, and keyboard shortcuts for power users. Each node displays its execution status, output preview, and error state directly on the canvas. The properties panel provides inline expression editing with autocomplete for the DSL syntax, model selection dropdowns, and file upload areas for skills and RAG documents.

Heym login page with animated workflow canvas background showing connected AI nodes

Authentication

Log into Heym

The login screen features an animated workflow canvas in the background, previewing the drag-and-drop editor before you even sign in. JWT authentication with HttpOnly cookies ensures secure sessions. Create your account with email and password to access the full dashboard with all 14 management tabs.

What's in the box

AI workflows that reason and run
Visible end to end

Animated Heym full feature set showcase
AI ASSISTANT

Describe it. Heym builds it.

Type or speak your automation in plain English — the Assistant generates the entire canvas: nodes, edges, credentials, expressions. Modify existing flows the same way: "add a HITL checkpoint before Output."

  • Works with any LLM credential you configure
  • Streams responses · auto-applies valid JSON
  • Voice input, transcribed in-browser
MULTI-AGENT

Agents that delegate to agents.

One orchestrator, five levels of nested sub-agents and sub-workflows. Each agent has its own model, tools, and Python execution context. Handoffs are visual — you can see the control flow, not infer it from logs.

  • Python tool calling per agent
  • MCP tool servers as drop-in tools
  • Auto context compression at 80% window
RAG · NATIVE

Vector stores, not another tool tab.

Upload PDFs, Markdown, CSVs directly. Managed Qdrant vector stores live alongside your workflows. The Qdrant RAG node wires retrieval into any flow with metadata filters and optional Cohere reranking.

  • One-click upload · chunking & embeddings handled
  • Hybrid search (dense + BM25 + rerank)
  • Team-shared knowledge bases
MCP · FIRST-CLASS

Your workflows become AI tools.

Every workflow is exposable over MCP. Paste the SSE URL into Claude Desktop, Cursor, or any MCP client. Your AI can now run your automations from its own interface. Agents can also consume any external MCP server as tools.

  • Bidirectional: MCP server + client
  • Per-workflow API keys
  • No SDK, no adapter code
HUMAN-IN-THE-LOOP

Approval checkpoints as a primitive.

Drop a HITL node and execution pauses, generates a public review link, sends it via your configured channel. A reviewer can accept, edit, or refuse. AI output never ships without someone accountable signing off.

  • Public shareable review link
  • Accept · edit · refuse · timeout
  • Reviewer history in the trace
OBSERVABILITY

851 traces on average per real alpha user.

Every LLM call, every agent step, every token and millisecond — captured automatically. Filter by source, credential, model. The trace viewer shows the full prompt, response, tool calls, and timing waterfall. Replay any run against a different model.

  • Span timeline per node execution
  • Data Pin · freeze outputs for testing
  • Built-in evals with LLM-as-Judge

Everything in One Dashboard

Heym's dashboard gives you a unified control center for workflows, credentials, vector stores, teams, analytics, evaluations, and more — all accessible from a single navigation bar with 15 dedicated tabs.

Each tab is purpose-built for a specific aspect of workflow management. From organizing automations in folders to inspecting LLM traces, from managing team permissions to running evaluation suites against multiple models — the dashboard eliminates the need for external tools and keeps your entire AI automation stack in one place.

Workflows

Create, organize, and manage your automation workflows in a card grid or list view with folder and sub-folder support. Import workflows by dragging JSON files, or export them for backup and sharing between instances.

Templates

Save and reuse entire workflows or individual node configurations as templates. Share templates with everyone or specific users and teams for consistent automation patterns across your organization.

Variables

Manage persistent global variables that survive across workflow executions. Store counters, configuration values, accumulated lists, and shared state accessible from any workflow via $global.variableName expressions.

Chat

A direct LLM chat interface for testing models, prototyping prompts, and asking questions without building a full workflow. Supports streaming, markdown rendering, inline images, and voice input.

Credentials

Securely manage API keys and secrets used by workflow nodes. All values are encrypted at rest with AES-256 Fernet and masked in the UI. Share credentials with individual users or entire teams.

Vector Stores

Create and manage Qdrant vector stores for RAG pipelines. Upload documents in multiple formats, manage content and sources, and share stores with team members for collaborative knowledge bases.

MCP

Configure Model Context Protocol integration to connect agents to external tool servers. Each workflow with an Agent node can expose its tools via MCP for Claude Desktop, Cursor, and other compatible clients.

Traces

Inspect LLM execution traces with full request and response payloads, timing breakdowns for LLM and tool calls, and skills included in each invocation. Essential for debugging agent behavior and optimizing prompts.

Analytics

View execution metrics including total runs, success rates, and average duration with period-over-period trend analysis. Filter by time range or specific workflow and track performance over 7, 30, or 90 day windows.

Evals

Create evaluation suites with test cases to systematically test AI workflows. Run evaluations against multiple models simultaneously, compare pass/fail rates, and use LLM-as-Judge scoring for automated quality assessment.

Teams

Create teams and add members by email to share workflows, credentials, variables, templates, and vector stores with a group at once. All team members automatically gain access to shared resources.

DataTables

Create structured data tables with typed columns including string, number, boolean, date, and JSON. Manage rows with inline editing, import and export CSV files, and share tables with read or write permissions.

Drive

Browse all files generated by skills across your workflows. Search, download, share, or delete files with full metadata including source node, creation date, file type, and size.

Scheduled

See all active cron workflows on a day, week, or month calendar. Each block shows the workflow name and cron expression — hover to preview the schedule, click to jump straight to the canvas.

Logs

View Docker container logs for the entire Heym stack including backend, frontend, and PostgreSQL. Filter by container and log level for efficient debugging and infrastructure troubleshooting.

Platform guides on GitHub
Click any card to view →

Feature guides on GitHub

Each card opens the canonical Markdown guide in the heymrun repository — AI assistant, agents, canvas, guardrails, portal, security, and more.

Reference

30+ Topics

Advanced reference documentation covering the AI assistant, agent architecture, canvas features, guardrails, parallel execution engine, expression DSL, credential sharing, portal chat UI, skills system, keyboard shortcuts, and security hardening.

AI Assistant

Natural language workflow builder inside the canvas editor. Describe what you want and the AI generates nodes and edges automatically. Supports voice input and streaming responses.

Agent Architecture

Deep dive into agent orchestration patterns including lead/sub-agent delegation, tool calling, MCP integration, skills system, and configurable execution limits.

Canvas Features

Visual editor capabilities including drag-and-drop node placement, edge routing, zoom controls, minimap, snap-to-grid, node pinning, and output inspection panels.

Guardrails

Content safety filtering for LLM and Agent nodes. Blocks violence, hate speech, sexual content, prompt injection, and other unsafe categories with configurable error routing.

Parallel Execution

DAG-based scheduling that automatically runs independent nodes concurrently in a thread pool. Downstream nodes start as soon as dependencies complete.

Human-in-the-Loop

Approval checkpoints that pause workflow execution and generate a public review link. A human reviewer can approve or reject the pending action before it continues.

Expression DSL

Reference upstream data with $nodeLabel.field syntax. Supports string interpolation, ternary operators, array methods, object access, and built-in helper functions.

Credentials and Sharing

Encrypted credential storage with AES-256 Fernet. Share API keys with individual users or entire teams while keeping secret values masked and secure.

Portal Chat UI

Turn any workflow with an Input and Output node into a public chat interface. Embed on websites or share a link for end-user self-service powered by your workflows.

Drive & Generated Files

Skill-generated files in the Drive tab: search, download, share links with optional password and expiry. Use the Drive node in workflows to delete files or update access constraints programmatically.

Skills System

Portable capability bundles consisting of SKILL.md instructions and optional Python tools. Drag and drop onto Agent nodes to extend context and toolbox.

Keyboard Shortcuts

Power user shortcuts for node selection, deletion, duplication, undo/redo, zoom, panning, and canvas navigation to speed up workflow creation.

Edit History

Track all changes to workflows with timestamps and user attribution. Restore previous versions and compare differences between workflow revisions.

Webhooks and Triggers

Configure JSON and SSE webhook endpoints, Telegram and Slack bot webhooks, cron schedules, IMAP mailbox polling, outbound WebSocket client triggers, and RabbitMQ consumers for event-driven automation.

Execution Tokens

Scoped JWTs that grant access to a single workflow's execute endpoint. Create short-lived or long-lived tokens directly from the cURL dialog, select one to embed it in the command, and revoke at any time without affecting your session.

SSE Streaming

Stream workflow execution over stream with execution_started, node_start, node_complete, and execution_complete events. The editor can generate ready-to-run cURL commands and per-node start messages.

User Settings

Personal configuration including theme, default LLM provider, API key management, notification preferences, and display options for the canvas editor.

Security

Security architecture overview covering JWT authentication, HttpOnly cookies, credential encryption, role-based access control, and deployment hardening recommendations.

Workflow Organization

Organize workflows with folders, sub-folders, tags, and pinned favorites. Search workflows with the command palette and filter by status or recent activity.

Browse Full Documentation on GitHub