MagureAcademy

Video lessons · 39 lessons

Browse Lessons

Filter the whole catalogue by track or level, or search it from the header.

39 lessons shown

How-toBeginner

Enterprise LLM Configuration and Model Cost Control

Connect a model provider to MagOneAI: name a config, store the API key in KeyVault, load the provider's models, set what each one is allowed to do, test it, and cap cost per token before you create it.

1:53

How-toProfessional

Secrets Management: Vault-Backed, Write-Only Keys

An API key pasted into a workflow travels wherever the workflow goes. Save it once into the vault instead, and from then on MagOneAI only ever shows a masked preview while workflows use it without seeing it.

1:38

How-toBeginner

Enterprise Tool Integration with MCP Servers

Agents can converse, but they cannot search the web or send mail on their own. MCP servers give them tools. Connect the Outlook server over OAuth, read its tool schemas, and prove a call works before any workflow depends on it.

2:37

How-toProfessional

External MCP Servers: Bring Your Own Tools

The platform tools cover the common cases, but your business runs on more. If a system already has an MCP server, paste one HTTPS URL, add the headers it expects, and MagOneAI discovers its tools for you.

1:34

How-toProfessional

OAuth Integrations: Tools That Act as the User

A shared service account can send an email, but not as you. OAuth lets a tool act on behalf of the person who signed in: register the OAuth app once for the organization, then each person connects their own account.

0:56

How-toBeginner

Building Your First Enterprise AI Workflow

Build the smallest useful workflow (Start, Agent, End) from an empty canvas: define an input field, give the agent a job, map data between nodes with the variable picker, and run it against a real question.

4:12

How-toBeginner

Agent Library: Standardized, Reusable Agents

Save a configured agent to the library once, with its instructions, tools, knowledge, model settings and output schema, then drop a copy into any later workflow instead of rebuilding it from blank.

1:05

How-toProfessional

Prompt Debugging and Execution Tracing

When an agent answers oddly, stop guessing. Open the execution and read the exact prompt that went out, every variable already filled in, alongside the raw output that came back.

1:01

How-toProfessional

Test Cases: Regression Testing AI Workflows

You improved a workflow this morning. Did you also break it? Save test cases with the output you expect, run them all before every publish, and download the report.

1:36

How-toProfessional

Version Control and One-Click Rollback

Your edits do not reach users until you publish. Label versions so they still mean something in three months, publish the one you want live, and roll back to a known good state in one click.

1:09

How-toProfessional

Export and Import: Portable AI Workflows

Another team wants the workflow you built, in another project. Export it as one file, import it where it is needed, and MagOneAI tells you exactly what still has to be wired up.

1:10

How-toProfessional

Scheduled Workflow Execution

A schedule is a trigger with the workflow's inputs already filled in, because there is nobody there to fill in the form. Set when a run happens, bake in the values it needs, and manage every schedule from one view.

1:26

How-toProfessional

Exposing AI Workflows as Secure APIs

Create a scoped API key so an existing workflow can be called from your own systems, store the secret the one time it is shown, and revoke it cleanly when that integration retires.

1:38

How-toBeginner

Hub Tour: Run Published Workflows Without Studio

The people who use a workflow are rarely the people who built it, and they should never have to see a canvas. In Hub, a published workflow is something you open, fill in and run.

1:12

How-toBeginner

Hub Inbox: Approve and Resume Paused Workflows

A workflow paused for approval is only useful if the approver can find it. The task lands in their Hub Inbox with the context they need, and one click carries the workflow on from where it stopped.

2:25

How-toProfessional

LLM Usage Quotas and Cost Governance

Usage is metered as it happens, not estimated. Set token limits at organization, project, use case or user level, see what enforcement looks like at run time, and handle the increase requests that follow.

1:23

How-toProfessional

Single Sign-On: Microsoft, Google and Oracle

Another platform should not mean another password. With SSO your people sign in with the identity they already use at work, and a new joiner gets access on their first sign-in.

0:41

Node ExplainedBeginner

START Node: Workflow Inputs and Validation

Open the Start node and declare what a workflow needs: pick the default model, choose how data moves between nodes, then define input fields (string, number, enum, array, object, image, media, file) with the constraints each type allows.

2:02

Node ExplainedBeginner

END Node: Structured Workflow Outputs

Define what a workflow returns: choose an output format, map exactly which fields reach the result, and decide whether strict mode limits the response to what you mapped or lets everything from the last node through.

0:48

Node ExplainedBeginner

AGENT Node: Persona, Knowledge and Tools

Everything else in a workflow moves data, calls systems or branches. Only the AGENT node reads, decides and writes in language: an instruction, a model, and whatever knowledge or tools you attach.

2:26

Node ExplainedProfessional

TOOL Node: Deterministic Tool Execution

Call a system or MCP tool with no model in the loop. Pick the tool, whether that is a database query, a web search or a calendar, watch the configuration change to match it, and set a timeout so a slow tool cannot stall the run.

1:34

Node ExplainedProfessional

API Node: Direct REST Integration

Call a third-party REST API straight from the canvas: method, URL, query parameters, headers and body, then the settings that keep it safe in production: timeout, max response size, redirects, and which status codes count as failure.

1:44

Node ExplainedProfessional

CODE Node: Sandboxed Data Transformation

Run sandboxed code instead of prompting for it. Declare an input model and an output model, then write the logic, reading from inputs and assigning to the output, for the parsing and transformation a language model has no business doing.

1:08

Node ExplainedProfessional

CONDITION Node: Rule-Based and AI-Based Routing

Branch a workflow three ways: compare a variable, evaluate an expression, or let a model sort the input into buckets you name. Includes match-per-route, for several branches out of a single node.

2:16

Node ExplainedProfessional

USE CASE Node: Reusable Workflows as a Single Step

When two workflows need the same check, the copies start to drift. Build it once and call it as a single step: the child's START inputs become the node's inputs, and its END mapping becomes the node's outputs.

1:38

Node ExplainedProfessional

PARALLEL Node: Concurrent Execution and Merge Strategies

Run agents at the same time instead of one after another. Give each its own branch, add conditions if a branch should only sometimes run, then choose a merge strategy for how the results come back together.

1:26

Node ExplainedProfessional

HUMAN TASK Node: Approval and Escalation Workflows

Pause a run and ask a person. Set up an approval or a clarification task, choose who it goes to and what context they see, then route the workflow on the answer that comes back.

1:25

Use CasesAdvanced

WAF False Positive Detection: Automated Security Triage

Pull the requests your web application firewall blocked, compare them against your platform's real endpoints, and let an agent decide which blocks are false positives, then email the findings to the people who can unblock them.

1:43

EssentialsBeginner

Platform Fundamentals

The mental model behind every MagOneAI workflow: the five nouns the platform comes down to, what happens when a workflow runs, and who builds in Studio and who runs in Hub.

9:31

EssentialsBeginner

Agent and Workflow Design

Most of the quality in an AI workflow lives in how you configure and instruct the agent. The canvas first, then the agent, with two labs to build along with.

12:11

EssentialsBeginner

Enterprise Tool Integration with MCP

A model can only produce text. Tools are how a workflow reaches the outside world, and one sentence explains most tool failures you will ever see: MCP servers need credentials.

6:30

EssentialsProfessional

Control Flow and Orchestration

Condition and Parallel look almost identical on the canvas and do different jobs. One is routing, where exactly one branch runs; the other is concurrency, where any number can.

5:08

EssentialsProfessional

Enterprise RAG and Knowledge Bases

A model does not know your pricing policy, your product manual or your board pack. A knowledge base grounds the agent in your documents, provided its instructions forbid answering beyond what was retrieved.

6:23

EssentialsProfessional

Production Deployment and Governance

The stretch from a clever prototype to something you would put your name on: files, approvals, dynamic values, tests, publishing, and what to do when a live run breaks.

17:58