What is ActionFlow

An ActionFlow is a visual, node-based workflow on the ActionFlows platform. You connect triggers, AI models, integrations, and human approval on a canvas. The graph runs without writing glue code.

You edit the graph in Actionflow Studio. LLM nodes, image generation, and transcription are catalog steps, not plugins. Named organization connections authenticate integrations. Estimated token cost appears on the studio bottom bar when cost calculation is enabled for the organization. Runs consume organization credits. See Usage & Cost.

Each node is one step: model, integration, control flow, storage, or HITL. You still pick connections, inputs, and wiring. The graph itself does not require code.

Who uses ActionFlows

• Teams who automate support, routing, and back-office graphs
• Developers who trigger those graphs from a backend (API, SDK, hosted MCP)
• Operators who need run history, HITL approvals, and credit usage in one organization

What an ActionFlow is made of

• One start node: Start, Start Scheduled, or Start Subflow
• Connected nodes: AI, control, integrations, storage, HITL
• Run inputs: empty required fields as {nodeId}_{inputName}
• Status: new flows are passive until Settings → Set to active

Inbound HTTP is a flow webhook on a Start graph, not a start-node type. Chat channels are agent triggers.

Where to go next

• Getting Started: build your first flow in 10 minutes
• Key Concepts: understand flows, runs, triggers, and nodes
• Agents: durable conversational AI (Studio and API), alongside flows
• Nodes: the full catalog of building blocks
• Trigger: how the graph starts
• Actionflow Studio: canvas, inspector, and run bar
• Settings: active status, visibility, and delete