Agentic AI Capabilities

Thirty-nine mechanisms in six sets, in the order you meet them.

Every capability of Lima Juno — Agentic AI has its own page: what it does, the steps it runs, what you see on screen, and the evidence behind each number.

THE SIX STAGES, IN THE ORDER YOU MEET THEM01Registerdata sets acrossyour applications,read where they live02Maptables and columnsto the supply chainontology03Askquestions in plainlanguage, answeredfrom live data04Teachyour vocabulary,your acronyms, yourcorrections05Actagents that draft orcarry out actions atyour level06Learnaccepts, edits andoverrides improvethe next cyclewhat your team confirms, corrects and approves returns to every stage
Each stage produces what the next one needs. The dashed line is the part people often miss: confirmations, corrections and approvals feed back into every stage.
39
capability pages, one for each mechanism in the product
product
6
sets: register, map, ask, teach, act, learn
product
40+
kinds of source system that can be registered
product
A0–A4
autonomy levels, set by you for each agent
product
Where to begin

The six stages, and the page that opens each one

The order matters: each stage produces what the next one needs, and trust grows in the same order.

Each stage links to its first page. The pill row at the top of every capability page carries the rest of that set.

Register · 4 pages

Your data stays in your systems, and Juno reads it there

Four pages on connecting a source, choosing tables, reading in place and keeping up with changes.

Map · 5 pages

Juno proposes what each column means, and your team confirms it

Five pages on the ontology, the proposals Juno makes, the review screen and linking the same thing across systems.

Ask · ChatLima · 8 pages

A question in plain language, an answer with its working on show

Eight pages on ChatLima: from a typed question to a checked answer, in the places where people already work.

Teach · 5 pages

Your words, your acronyms and your feedback, kept in one place

Five pages on teaching Juno your vocabulary, keeping its grounding current and measuring its accuracy.

Act · 9 pages

Policies written in plain language, executed the same way every run

Nine pages on the task agents, the plan behind them, the rules engine, approvals and confirmed write-back.

Learn and govern · 8 pages

Better each cycle, and accountable from signal to acknowledgement

Eight pages on the learning loop, the audit trail, versions, cost control and the manager’s view.

Where to start

The three pages that answer your first question, by the chair you sit in

Six starting points; each one takes about fifteen minutes to read.

Supply plannerBegin with ChatLima, then question paths for the Monday review, then the ten task agents.
BuyerBegin with recommendations, then approvals, then write-back to the ERP.
Data and analytics leadBegin with data registration, then automatic mapping, then the knowledge index.
IT architectBegin with data in place, then security, then your choice of model.
Supply chain directorBegin with the task agents, then the autonomy ladder, then observability.
Finance, risk and auditBegin with the audit trail, then versioning, then secure answers.
How to read these pages

The same shape on every page, so a claim is quick to check

Figures carry the mechanism; the prose carries the steps; the labels carry the evidence.

Each capability page opens with a figure of the mechanism, then a short band of figures with their evidence labels. A “step by step” section follows the work in order, and the sections after it show the screens, the tables and the interfaces behind it.

The running example runs through the whole set: one planner, 1,240 exception messages on a Monday, and a cancellation window that closes on Thursday. Each page carries the next instalment and a link to the page that follows.

Every page closes with the terms it used, the questions buyers ask, and internal review notes that say where each claim comes from and who still has to confirm it.

Five labels for every numberProduction: measured in a live estate. Anonymised engagement: measured with a client who stays unnamed. Showcase estate: our demonstration estate. Estimate: a calculation. Dated source: a published fact with its date.
One running exampleThe same planner, the same 38 purchase order lines and the same $412,000 appear on every page, so each mechanism has something concrete to hold on to.
Analyst questions welcomeWhere a capability is still being built, the review notes on that page say so. Mechanisms, interfaces and limits are described in the detail a briefing needs.
Questions
Which pages should an analyst read first?

Read automatic mapping, show the working, the rules engine, approvals and the audit trail. Those five cover meaning, grounding, execution, oversight and evidence.

Do we have to adopt all six stages?

Each stage is useful on its own. Registration and mapping give ChatLima; the act and learn stages build on the same ontology and access rules when you are ready.

Where do the numbers on these pages come from?

Every figure carries one of five labels: production, anonymised engagement, showcase estate, estimate or dated source. The label sits beside the number.

In the platform

Where Juno sits among the four capabilities

Each one builds on the one before it.

One shared foundationCatalogLineageAccess policyWrite path

See Juno answer and act on your own data

Bring two systems and one decision your planners make every week. In the first week we register the data, confirm the mapping with your team, answer your questions in ChatLima and run the first agent at the level you choose.