Five end-to-end chains run a manufacturer, and in each the market is deploying three or four agentic use cases now.
The chains differ; the parts an agent is configured from, and the week that builds one, are the same.
Lima’s agent blueprint is a design of 960 agent roles across five end-to-end chains. It is the map of what Juno can be configured to do, and the first agents are built from it one at a time, the way the PO cancellation agent was.
Every use-case page follows one shape: a moment in the day, the evidence, the six stages, who does what at which level, the settings you would make and the roles it would draw on. “Would” is the honest tense on nineteen of the twenty pages; the twentieth describes the agent that runs. blueprint
One line on each chain from its research note, and the use cases it lights.
Most of the cost sits in people handling documents and exceptions.
Open the chain →The market measures it in cost per document.
Open the chain →Analysts find the most AI in use here, with 74% of enterprises reporting AI in S&.
Open the chain →The close has barely moved in a decade: 59% of companies close within six business days, against 60% in 2019 (ISG, 12 December 2023), and reconciliations and accruals take the time.
Reconciliations and close tasks · Journal entry anomalies · Accruals for goods received · Flux and variance commentary
Verification takes 60% to 70% of a chip project’s effort (Siemens, 3 September 2025), and first-silicon success fell to 14% in 2024 (Semiconductor Engineering, 27 March 2025).
Regression triage and debug · Engineering change impact · Component end of life · Launch readiness checks
The chain page carries the market evidence, the four use-case pages and the additional capability by phase. Sources at the foot of the page.
Four pages per chain; the one labelled production is the PO cancellation agent of chapter 5.
The additional roles are designed and yet to be built; each chain page lists them by phase. blueprint
The chair you sit in decides the first use case; the week runs the same way for all of them.
| Role | First use case | Why start there |
|---|---|---|
| Supply planner (L. Tan) | Purchase order change, cancellation and confirmation chasing | The one agent that runs. Her policy sentence is the starting point; her Monday shrinks from 2.5 days to 38 minutes on the showcase estate. |
| Buyer (D. Okafor) | Supplier risk, sanctions and end-of-life monitoring | Supply continuity is the top procurement priority for 2026 (The Hackett Group, 17 March 2026); the roles would start at A1 and A2. |
| Finance lead (J. Harrison) | Cash application and remittance matching | The median company auto-matches 70% of payments (The Hackett Group, 11 August 2025); the A4 scene of the gamut, audited after the fact. |
| Controller | Account reconciliations and close task automation | 55% of accounting teams name reconciliations their top automation opportunity, while 5% call them highly automated (FloQast, 11 August 2026). |
| Data lead (A. Novak) | Engineering change impact across BOM, stock and orders | One change touches PLM, two ERPs, the planning suite and the supplier portal; the map across them is hers to build in the first week. |
| IT architect (S. Iyer) | Invoice matching and exception clearing | The best-measured cost per document in any chain; write-back through the ERP’s released interfaces after a confirmed test write. |
Whichever row is yours, the first week is the same: two source systems, one weekly decision, three people for two hours a day. The use case sets which tables are registered on the Monday, which question is asked on the Wednesday and which policy is written on the Friday.
One does: purchase order change and cancellation, in production at a high-tech manufacturer. The other nineteen are designed on the blueprint; each would be configured from the same parts, simulated first and released at a level a manager signs.
Bring two source systems, one weekly decision and three people for two hours a day. 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.
