TurfAI User Guide

This guide is for the product team — the people who need the most complete picture of what TurfAI does, how each module behaves, and how it is administered and deployed. It is deliberately visual: lean on the diagrams to build intuition before you dive into the tables.

How to read this guide

The guide is organised from concepts → modules → reference, so you can start anywhere:

• Concepts — the mental models. What agentic automation is, and when to reach for a workflow vs. an agent vs. a squad. Read this first if the vocabulary (RAG, HITL, ReAct, blackboard, governance) is new to you.
• Modules — one page per product module, all built from the same 9-section template (What it is · When to use it · How it works · Sub-features · Configuration · Dependencies · Limitations · How to test · APIs used). Because every page is structured identically, you can scan the same section across modules.
• Features — the capability surface at a glance, grouped by area.
• Node catalog — every task type you can drop on the canvas, by category, with its config parameters.
• Administration — ownership, RBAC, seeding, usage & analytics.
• Deployment — how TurfAI is run in production.

The platform at a glance

TurfAI is an enterprise workflow automation platform powered by agentic AI. Users build workflows on a visual canvas; each workflow is triggered (webhook, schedule, or event), runs a graph of tasks, agents, and squads, can pause for human review, and then acts on external systems — with governance (RBAC, audit, Data Shield) around the whole loop.