Skip to main content

​
What Is a Workflow?

A workflow is the step-by-step plan your agent follows to complete a task. When you describe what you want and hit Build, Decisional generates a workflow — a directed graph of nodes connected by edges. Each node represents one step in the process. Edges define the order of execution.

​
Node Types

Workflows are made up of different node types, each handling a specific kind of work:

Trigger Node

The starting point of every workflow. Defines how the agent gets activated — manual run, schedule, webhook, or an event from a connected app.

Task Node

The core workhorse. Reads data, processes information, makes decisions, and produces output. Most of your workflow logic lives here.

Action Node

Takes an action in an external app — sends an email, posts a Slack message, updates a CRM record, writes to a database. Action nodes are flagged during testing so you can approve before they execute.

AI Node

Runs an AI model to process, classify, or summarize data. Use this when you need the agent to reason over content, categorize items, or generate text.

RAG Node

Searches your uploaded sources using Retrieval-Augmented Generation. Pulls the most relevant chunks from your documents so other nodes can use that context.

​
Viewing Your Workflow

After your agent builds, click the Workflow tab to see the visual graph. Each node is displayed with:
  • Its name and a short description of what it does
  • Its type (trigger, task, action, AI, RAG)
  • Connections showing the flow of data between steps
You can zoom, pan, and click on any node to see its details and the code it runs.

​
AI Nodes

AI Nodes let your agent call an AI model as a step in the workflow. This is useful for:
  • Classification — categorize emails, tickets, or documents into types
  • Summarization — condense long content into key points
  • Extraction — pull structured data from unstructured text
  • Generation — write responses, draft emails, or create content
The AI Node takes an input (text, data, or context from previous nodes) and produces an output that flows to the next step.

​
RAG Nodes

RAG Nodes connect your workflow to your uploaded Sources. When a RAG Node runs, it:
  1. Takes a query (from the current context or a previous node)
  2. Searches your indexed documents for the most relevant chunks
  3. Returns the matching content with citations
This lets your agent pull real information from your documents mid-workflow — pricing from a price list, terms from a contract, procedures from an SOP.
RAG Nodes work best when your sources are well-organized and your agent’s instructions reference specific documents by name.

​
How Workflows Get Built

You don’t need to design workflows manually. Here’s the process:
  1. You describe what you want in plain English
  2. Decisional generates the workflow automatically — choosing the right node types, wiring them together, and writing the logic
  3. You review the workflow in the visual viewer
  4. You can modify by telling your agent in the chat what to change
If you need to make changes after building, just tell your agent — for example, “add a step to send a Slack notification after the email is sent” — and it will update the workflow.

​
Action Node Approvals

Action nodes are special because they affect external systems (sending emails, updating CRMs, posting messages). During testing:
  • Your agent pauses before executing each action node
  • You see a preview of what the action will do
  • You can approve or reject each action individually
This gives you full control before anything touches your real data or apps.
When your agent is live, action nodes execute automatically unless you’ve configured human-in-the-loop approval in your instructions.