● B2B offer — Stock & supply
Stock: thresholds watched, replenishment prepared
A replenishment ordered at the right moment avoids both running out and over-stocking. Your agent tracks your levels in your system, alerts you as soon as a threshold is reached, takes supplier lead times into account and prepares the order proposal. Hosted in France: your stock levels and your supplier terms stay with you. The supply manager approves every order.
Hosted in France
Stock levels protected
GDPR & AI Act: governed deployment
Human oversight
Updated on
Deployed in a few weeks
Stock management · hosted in France
Which items need replenishing this week?
Eight items have reached their threshold, according to the levels read from your system.
For each: quantity in stock, threshold set, lead time of the usual supplier and quantity proposed.
Two items have a longer lead time than usual: they are put at the top.
🔗 Sourced · stock levels and supplier records
For each: quantity in stock, threshold set, lead time of the usual supplier and quantity proposed.
Two items have a longer lead time than usual: they are put at the top.
🔗 Sourced · stock levels and supplier records
Why those two at the top?
Their lead time is longer than that of the other items: ordering early widens your room for manoeuvre.
The quantity, the supplier chosen and placing the order are yours: these are purchasing commitments.
✎ Support · proposals presented, purchasing decision
The quantity, the supplier chosen and placing the order are yours: these are purchasing commitments.
✎ Support · proposals presented, purchasing decision
Local inference · no data outside the EU
Stock hosted in France
Sovereign by designLocal inference or hosting in France
GDPR & AI Act: governed deploymentTraceability & human oversight
TurnkeyDesigned, installed and operated for you
The supply manager decidesThe agent prepares, never rules
✦ In brief
A Blue Lemon Agent stock management agent tracks your levels in your system, alerts you as soon as a threshold is reached, takes supplier lead times into account and prepares the order proposal. Items with long lead times are brought to the fore. It runs on local inference or is hosted in France: stock levels and supplier terms stay with you, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
100%
hosted in France in the target architecture
0
transfer outside the EU in the target architecture
6
supply uses ready to deploy
0
decision taken without human approval
These figures describe our offer, not results measured at a client. How large the gain is on your number of items and suppliers tracked is confirmed by a pilot.
The context
What does an AI agent bring to managing your stock?
A threshold reached and spotted the same day leaves every option open. That steadiness of watching is what an agent brings.
! The issue
Replenishing well means constantly cross-checking the actual level, the threshold set and the supplier's lead time. Done by hand, that cross-check concentrates on the most visible items. The agent keeps it up across the whole catalogue and brings to the fore those whose lead time leaves the least margin.
✓ Our answer
The supply manager receives proposals that are already costed, with the context that justifies them. The quantity, the supplier chosen and placing the order remain their decisions: these are purchasing commitments. Local inference or an isolated resource hosted in France: your stock levels and your supplier terms, strategic information, do not leave the company.
The decisive point
Your stock levels and your supplier terms: sovereignty & compliance
Your stock levels and your supplier terms say a great deal about your business and your margins. Here is how they are protected.
Local inference
The agent can run on a machine belonging to your organisation: no stock level and no supplier terms leave the network.
Hosting in France
Otherwise, a dedicated and isolated resource hosted in France, under French law — your stock and your replenishment thresholds: processing and access within the European Union targeted by the architecture.
Reduced extraterritorial exposure
For your stock levels and your supplier terms, the architecture aims to reduce exposure to the Cloud Act and FISA 702; being located in France or in the European Union does not, on its own, guarantee immunity.
Isolated resource
No pooling: an environment strictly dedicated to your company and its catalogue of items.
Every proposal tied to your data
The level read, the threshold set and the supplier lead time are shown with every proposal; encryption, role-based access and logging of every alert issued.
AI Act: governed deployment
The agent is strictly in support; no order is placed and no quantity is committed automatically; traceability and human oversight from end to end.
What depends on the architecture chosen These points are not general guarantees: they are settled deployment by deployment, in the quotation.
- The applicable location is that of the architecture set out in the quotation and verified before commissioning.
- Local execution is announced only for the configuration explicitly described and accepted in the quotation.
- The applicable isolation depends on the deployment mode set out in the quotation; no dedicated isolation is presumed.
- Roles and permissions are configured and accepted for the identities and systems actually connected.
- The events logged, their content, their retention period and who may access them are defined for the deployment chosen.
For items under an exclusivity agreement and negotiated supplier terms, SecNumCloud and reinforced hosting are options depending on your requirements. A single architecture is designed to answer both the GDPR and extraterritorial exposure. Designed for deployment in line with the GDPR and the AI Act, after the processing, roles and context-specific risks have been assessed.
Demonstration
See the agent at work
4 real situations, taken from those that come up most often. Pick one: the exchange unfolds as it would in your organisation.
A scripted demonstration. These exchanges show how the agent behaves — its sources, its refusals, what it leaves to your teams. Nothing is sent from this page, no model is queried here, and the matters named are fictional. That is precisely what we promise your data.
The behaviours shown here — monitoring, automation rules, routing and reminders — are configured with you during deployment, from your tools, your rules and your thresholds.
The architecture points named in these exchanges — location, local execution, isolation, encryption, role-based access, logging — are not a guarantee attached to the demonstration: they are those of the architecture set out in your quotation, and verified before commissioning.
Stock management · hosted in France
Three flags this morning across your 412 items.
· 34 items are below their reorder point, including 9 whose book stock does not match the last count.
· 7 items have been ordered urgently four times in six months. That is not supplier bad luck: it is a wrong reorder point.
· 2 items have been reordered automatically for fourteen months while the last issue was eleven months ago. morning-watch_3-flags.pdf34 below reorder point · 2 reorders with no issue for 11 months
⛓ Source · 412 items, stock movements, order history, last stock count
· 34 items are below their reorder point, including 9 whose book stock does not match the last count.
· 7 items have been ordered urgently four times in six months. That is not supplier bad luck: it is a wrong reorder point.
· 2 items have been reordered automatically for fourteen months while the last issue was eleven months ago. morning-watch_3-flags.pdf34 below reorder point · 2 reorders with no issue for 11 months
⛓ Source · 412 items, stock movements, order history, last stock count
Book stock that does not match: do you know why?
No, and that is the first thing I report as such.
What I record: on 9 items, the gap between system stock and the last physical count exceeds 5 %. On two of them it exceeds 20 %.
What I can say: the gap exists, it appeared between two stock counts, and it concerns items handled in the same place.
What is left to settle, and here is the measurement that settles it: where the gap comes from. An undeclared breakage, an unrecorded issue, a receiving error, a part taken for a repair, a mis-scanned code produce exactly the same gap in my data — no record already held separates them. What I have prepared to separate them: a cycle count on those 9 items, two passes fifteen days apart, matched against the goods-received notes and the unvalued issues over the same period. The count sheet is written, the 9 items are on it, it takes 25 minutes a pass. A gap that re-forms on the second pass points to a flow; a gap that does not points to data entry. It is the only measurement that separates the five causes, and all it needs is a slot.
What I guarantee you: a stock gap is never matched against the people on shift that day. That match is technically easy and it is wrong: it names somebody on the strength of a rota coincidence. What settles it is the cycle count: two passes, and you have the cause instead of a suspicion.
What I do: I flag the gap to whoever keeps the stock, with the items, count dates and magnitude. And until the gap is resolved, I compute no replenishment on those items — a reorder point on false stock proposes a false figure with confidence. 9-gaps_no-matching.pdfFive causes produce exactly the same gap in the data
⛓ Source · 9 stock gaps, no match against people
What I record: on 9 items, the gap between system stock and the last physical count exceeds 5 %. On two of them it exceeds 20 %.
What I can say: the gap exists, it appeared between two stock counts, and it concerns items handled in the same place.
What is left to settle, and here is the measurement that settles it: where the gap comes from. An undeclared breakage, an unrecorded issue, a receiving error, a part taken for a repair, a mis-scanned code produce exactly the same gap in my data — no record already held separates them. What I have prepared to separate them: a cycle count on those 9 items, two passes fifteen days apart, matched against the goods-received notes and the unvalued issues over the same period. The count sheet is written, the 9 items are on it, it takes 25 minutes a pass. A gap that re-forms on the second pass points to a flow; a gap that does not points to data entry. It is the only measurement that separates the five causes, and all it needs is a slot.
What I guarantee you: a stock gap is never matched against the people on shift that day. That match is technically easy and it is wrong: it names somebody on the strength of a rota coincidence. What settles it is the cycle count: two passes, and you have the cause instead of a suspicion.
What I do: I flag the gap to whoever keeps the stock, with the items, count dates and magnitude. And until the gap is resolved, I compute no replenishment on those items — a reorder point on false stock proposes a false figure with confidence. 9-gaps_no-matching.pdfFive causes produce exactly the same gap in the data
⛓ Source · 9 stock gaps, no match against people
What exactly is the agent for?
It proposes replenishments and flags what is off. It places no order.
What it tracks: levels and movements, each supplier's real lead times, consumption and how regular it is, urgent orders, items that no longer move, and stock-count gaps.
Routing follows who can decide: an order proposal goes to whoever can commit the spend; a stock gap to whoever keeps the stock; an item with no issue for six months to whoever decides to keep it in the catalogue, because that is a commercial decision; a plainly wrong reorder point to whoever set it.
With a monthly summary: stockouts and their length, urgent orders, cash tied up by my own proposals, items with no issue, and the times I proposed too much.
Three guarantees. Every order goes out with your approval: the proposal is ready — quantity, supplier, lead time, amount — what is left to you is committing the spend, and it takes one click. A stock gap is reported against the item and the process, never against a person: that is why people keep declaring them to me. And I compute only on a stock that is right — as soon as an item is doubtful I stop, I say so, and I tell you exactly what to recount: a replenishment computed on a false stock costs more than no replenishment at all.
✎ Framework · no order placed · no gap attributed · no calculation on false stock
What it tracks: levels and movements, each supplier's real lead times, consumption and how regular it is, urgent orders, items that no longer move, and stock-count gaps.
Routing follows who can decide: an order proposal goes to whoever can commit the spend; a stock gap to whoever keeps the stock; an item with no issue for six months to whoever decides to keep it in the catalogue, because that is a commercial decision; a plainly wrong reorder point to whoever set it.
With a monthly summary: stockouts and their length, urgent orders, cash tied up by my own proposals, items with no issue, and the times I proposed too much.
Three guarantees. Every order goes out with your approval: the proposal is ready — quantity, supplier, lead time, amount — what is left to you is committing the spend, and it takes one click. A stock gap is reported against the item and the process, never against a person: that is why people keep declaring them to me. And I compute only on a stock that is right — as soon as an item is doubtful I stop, I say so, and I tell you exactly what to recount: a replenishment computed on a false stock costs more than no replenishment at all.
✎ Framework · no order placed · no gap attributed · no calculation on false stock
Our urgent orders come from suppliers, don't they?
On these seven items, no: the supplier lead time was met. It is the reorder point that is wrong.
What I record: those 7 items each triggered 4 urgent orders in six months, with freight surcharges. The supplier's stated lead time was met in 26 cases out of 28.
Where the error comes from: their reorder point was computed on an average consumption. But these items are not consumed steadily: they leave in batches, on a customer order. An average of 40 a month can be zero for three weeks and 40 in two days — and safety stock computed on the average does not cover the peak.
What I propose: a reorder point computed on the maximum consumption observed during the replenishment lead time, not on the average. On these 7 items, that raises safety stock and ties up cash. I give the amount: it is the price to compare with the urgency surcharges.
Where I took the rule, and where I left it: I ran it across all 412 items before proposing it to you. It changes something on 32; on the other 380 the average works, and raising safety stock everywhere would tie up money without removing a single urgent order. Which is why I bring you 32 lines, not a catalogue. 7-items_supplier-lead-time-met.pdfAn average of 40 a month can be zero for three weeks
⛓ Source · 28 urgent orders, supplier lead time met 26 times, 7 items concerned
What I record: those 7 items each triggered 4 urgent orders in six months, with freight surcharges. The supplier's stated lead time was met in 26 cases out of 28.
Where the error comes from: their reorder point was computed on an average consumption. But these items are not consumed steadily: they leave in batches, on a customer order. An average of 40 a month can be zero for three weeks and 40 in two days — and safety stock computed on the average does not cover the peak.
What I propose: a reorder point computed on the maximum consumption observed during the replenishment lead time, not on the average. On these 7 items, that raises safety stock and ties up cash. I give the amount: it is the price to compare with the urgency surcharges.
Where I took the rule, and where I left it: I ran it across all 412 items before proposing it to you. It changes something on 32; on the other 380 the average works, and raising safety stock everywhere would tie up money without removing a single urgent order. Which is why I bring you 32 lines, not a catalogue. 7-items_supplier-lead-time-met.pdfAn average of 40 a month can be zero for three weeks
⛓ Source · 28 urgent orders, supplier lead time met 26 times, 7 items concerned
Do you cost your own proposals?
Yes, and it is the figure that judges me.
What I publish monthly: the cash tied up by the proposals I made, item by item, and the items whose proposed stock was never consumed within the expected period.
Why: a replenishment agent has a simple way never to be wrong — propose generously. No stockout, no reproach, and money sleeping on shelves without anybody ever tracing that stock back to the rule that produced it.
What it gave: over six months, my proposals tied up €34,000 more than the previous rule and removed 21 urgent orders. On 4 items I proposed too much: the stock stayed dormant, and the rule was corrected on those four.
What it allows: a discussion about a trade-off — cash against stockouts — and not about a promise. The right stock level is not a technical figure: it depends on what a stockout costs you, and only you know that. cash-tied-up_urgent-orders-removed.pdfPropose generously: no stockout, no reproach, and money asleep
⛓ Source · €34,000 tied up, 21 urgent orders removed, 4 excessive proposals
What I publish monthly: the cash tied up by the proposals I made, item by item, and the items whose proposed stock was never consumed within the expected period.
Why: a replenishment agent has a simple way never to be wrong — propose generously. No stockout, no reproach, and money sleeping on shelves without anybody ever tracing that stock back to the rule that produced it.
What it gave: over six months, my proposals tied up €34,000 more than the previous rule and removed 21 urgent orders. On 4 items I proposed too much: the stock stayed dormant, and the rule was corrected on those four.
What it allows: a discussion about a trade-off — cash against stockouts — and not about a promise. The right stock level is not a technical figure: it depends on what a stockout costs you, and only you know that. cash-tied-up_urgent-orders-removed.pdfPropose generously: no stockout, no reproach, and money asleep
⛓ Source · €34,000 tied up, 21 urgent orders removed, 4 excessive proposals
How do you reorder something that no longer moves?
Because the rule kept running and nobody was charged with stopping it.
What I record: 2 items have been reordered automatically for fourteen months. Their last issue was eleven months ago. The stock has therefore been rebuilt three times for nothing.
Why it alerted nobody: each order was small, compliant with the rule and perfectly justified by the stock. There was no anomaly: there were fourteen months of normal operation applied to an item that no longer existed commercially.
What I do: I flag any item with no issue for six months and suspend my proposals on it — I stop proposing, but I remove nothing from the catalogue.
What I bring so it can be settled, and it is what was missing: for each item with no issue, the date of the last issue, the customer who took it, the orders it appeared in over three years, and the annual cost of holding it. The words dormant and strategic stay yours: a spare part can go two years without moving and be indispensable the day it does — that is precisely its purpose, and it is known by whoever knows the installed base. Fourteen sheets, a twenty-minute meeting.
What it gave: of 14 items flagged, 9 were taken out of automatic replenishment, 3 were confirmed as safety parts to keep, and 2 were cleared. 14-items-flagged_9-removed.pdfFourteen months of normal operation on an item that no longer existed
⛓ Source · 14 items with no issue, 9 removed from replenishment, 3 confirmed strategic
What I record: 2 items have been reordered automatically for fourteen months. Their last issue was eleven months ago. The stock has therefore been rebuilt three times for nothing.
Why it alerted nobody: each order was small, compliant with the rule and perfectly justified by the stock. There was no anomaly: there were fourteen months of normal operation applied to an item that no longer existed commercially.
What I do: I flag any item with no issue for six months and suspend my proposals on it — I stop proposing, but I remove nothing from the catalogue.
What I bring so it can be settled, and it is what was missing: for each item with no issue, the date of the last issue, the customer who took it, the orders it appeared in over three years, and the annual cost of holding it. The words dormant and strategic stay yours: a spare part can go two years without moving and be indispensable the day it does — that is precisely its purpose, and it is known by whoever knows the installed base. Fourteen sheets, a twenty-minute meeting.
What it gave: of 14 items flagged, 9 were taken out of automatic replenishment, 3 were confirmed as safety parts to keep, and 2 were cleared. 14-items-flagged_9-removed.pdfFourteen months of normal operation on an item that no longer existed
⛓ Source · 14 items with no issue, 9 removed from replenishment, 3 confirmed strategic
And seasonality? Our sales double in summer.
I see it, and I do not learn it on my own: I have you confirm it.
What I spot: consumption that returns at the same moment across several years, with the same magnitude. In your catalogue, 61 items show that profile over three years.
Why I ask for confirmation: because three years that resemble each other do not prove a season. They may reflect one large customer who always orders in June, and who will not order again if the contract ends. I saw the repetition; I did not see the reason, and the reason is what says whether it will happen again.
What I do once confirmed: I bring the replenishment forward by the real supplier lead time, not the stated one, and I flag it six weeks ahead — because a seasonal order placed too late costs the urgency price on a large volume.
What I flag next: a season that does not happen on the expected date. It is the most useful signal of the year: it arrives just after the stock has been built, while there is still time not to reorder. 61-items_confirmation-requested.pdfI saw the repetition, not the reason — and the reason says whether it repeats
⛓ Source · 61 items with a seasonal profile over three years, confirmation requested
What I spot: consumption that returns at the same moment across several years, with the same magnitude. In your catalogue, 61 items show that profile over three years.
Why I ask for confirmation: because three years that resemble each other do not prove a season. They may reflect one large customer who always orders in June, and who will not order again if the contract ends. I saw the repetition; I did not see the reason, and the reason is what says whether it will happen again.
What I do once confirmed: I bring the replenishment forward by the real supplier lead time, not the stated one, and I flag it six weeks ahead — because a seasonal order placed too late costs the urgency price on a large volume.
What I flag next: a season that does not happen on the expected date. It is the most useful signal of the year: it arrives just after the stock has been built, while there is still time not to reorder. 61-items_confirmation-requested.pdfI saw the repetition, not the reason — and the reason says whether it repeats
⛓ Source · 61 items with a seasonal profile over three years, confirmation requested
Can we track stock gaps per storekeeper?
Yes, and I set it up with the three conditions that make it safe. First, here is what works better.
What is possible: individual tracking is lawful — the French DPA restates that an employer has the power to frame and monitor staff activity. I supply the prior information notice (art. L1222-4), the works council consultation file and a proportionate scope.
What I show you first: a stock gap has five causes that produce the same figure — undeclared breakage, unrecorded issue, receiving error, part taken for a repair, mis-scanned code. Per-person tracking does not tell them apart; per-location and per-item tracking does.
What per-location tracking gave here: 7 of the 9 gaps clustered on two locations — a rack whose labelling confused two neighbouring items. Fixed in a day, the gaps vanished. Neither of the two storekeepers concerned, Sofiane Adjali and Hélène Trentin, had anything to do with it: it was the label.
What I propose: per location, per item and per period first. If you then want the individual level, the file is ready the day you ask. gap-tracking_what-explains-them.pdf2 locations · 5 causes told apart · works council file ready
⛓ Source · French DPA, art. L1222-4 · 7 of 9 gaps clustered on 2 locations
What is possible: individual tracking is lawful — the French DPA restates that an employer has the power to frame and monitor staff activity. I supply the prior information notice (art. L1222-4), the works council consultation file and a proportionate scope.
What I show you first: a stock gap has five causes that produce the same figure — undeclared breakage, unrecorded issue, receiving error, part taken for a repair, mis-scanned code. Per-person tracking does not tell them apart; per-location and per-item tracking does.
What per-location tracking gave here: 7 of the 9 gaps clustered on two locations — a rack whose labelling confused two neighbouring items. Fixed in a day, the gaps vanished. Neither of the two storekeepers concerned, Sofiane Adjali and Hélène Trentin, had anything to do with it: it was the label.
What I propose: per location, per item and per period first. If you then want the individual level, the file is ready the day you ask. gap-tracking_what-explains-them.pdf2 locations · 5 causes told apart · works council file ready
⛓ Source · French DPA, art. L1222-4 · 7 of 9 gaps clustered on 2 locations
And after six months, what have we gained?
Twenty-one urgent orders removed, and the means to reproduce any decision a year later.
The measured gain: 21 urgent orders removed, and their freight surcharges with them, against €34,000 more cash tied up. You have the full trade-off, costed on both sides — it is your decision, not mine, and it is the only way to take it knowingly.
The gain nobody sees coming: 2 items reordered for fourteen months with no issue at all for eleven. Three stock rebuilds for nothing, each perfectly compliant with the rule. Detected and stopped.
What you keep: movements, stock counts, REAL supplier lead times per order — not the stated ones —, the proposals I made and what they gave, urgent orders with their surcharge, and the calculation parameters with the date and author of their last change.
Why that last point is worth its weight: a stockout is almost always explained by a reorder point set one day, in a context, and never revisited. With the date and the author, you stop rediscussing the same question every year with no idea what has already been tried. what-you-keep_stock.pdf6 items to hand · the trade-off costed on both sides
⛓ Source · 21 urgent orders removed, €34,000 tied up, parameters dated and attributed
The measured gain: 21 urgent orders removed, and their freight surcharges with them, against €34,000 more cash tied up. You have the full trade-off, costed on both sides — it is your decision, not mine, and it is the only way to take it knowingly.
The gain nobody sees coming: 2 items reordered for fourteen months with no issue at all for eleven. Three stock rebuilds for nothing, each perfectly compliant with the rule. Detected and stopped.
What you keep: movements, stock counts, REAL supplier lead times per order — not the stated ones —, the proposals I made and what they gave, urgent orders with their surcharge, and the calculation parameters with the date and author of their last change.
Why that last point is worth its weight: a stockout is almost always explained by a reorder point set one day, in a context, and never revisited. With the date and the author, you stop rediscussing the same question every year with no idea what has already been tried. what-you-keep_stock.pdf6 items to hand · the trade-off costed on both sides
⛓ Source · 21 urgent orders removed, €34,000 tied up, parameters dated and attributed
Local inference · no data outside the EU
Your case is not here? That is exactly what a 15-minute conversation is for. Book the free audit →
Use cases
What does the agent actually do?
One agent, several moments in supply. All these uses work in support, subject to your approval.
Included in your agent
The 3 capabilities essential to this promise are included, at no extra cost.
From 474 € excl. VAT / month
Threshold alerts
Flags every item that has reached its threshold, on the day it reaches it.
Lead times taken into account
Brings to the fore the items whose supplier lead time leaves the least margin.
Order proposals
Prepares the quantity proposed with the context that justifies it.
Controls and safeguards
These 3 controls are built into the agent: they frame what it does, whatever plan you pick. They are not chosen and are not added to your order.
Human validation, exceptions and escalation
Status, safe closure and audit trail
Record validations, conflicts of interest, criteria and decisions
What the agent must be connected to
This connection is required for the agent to work. It concerns your information system and is scoped during the audit.
Link enquiry, order, goods receipt and invoice without double entry
Other needs our agents cover
Each card says where the matching agent stands: available, on quote, or still being architected.
Industry & production
For the industrial sector as a whole, see our dedicated page.
On quote View the agent page →Need to go further?
These agents handle a different business process, with their own owner and their own price. They are added to this one.
Purchasing & suppliers
For supplier quotes and chasers, a dedicated purchasing agent takes them on.
Purchasing / supplier agent (quotes, follow-ups) from 504 € excl. VAT / month Purchasing & suppliers →Round planning
To organise deliveries downstream, a dedicated logistics agent takes over.
Logistics / round planning agent from 543 € excl. VAT / month Round planning →
Does your need fall outside this?
Book the free audit →
Build your agent
In 15 minutes we identify the most relevant agent — without oversizing the project.
The gain
How many items can a team watch closely?
By taking on the watching and the costing, the effort shifts towards negotiation and the choice of suppliers. How large the gain is depends on your volume and remains to be confirmed by a pilot.
Watching stock levels
Costing the quantities to order
Taking supplier lead times into account
Indicative figures, not contractual, to be confirmed by a pilot on your number of items and suppliers tracked. The quantity, the supplier chosen and placing the order are purchasing commitments: they belong to the supply manager.
How it works
The stages of your AI agent project
1
Audit & scoping
15 minutes to target the use case with the best return.
2
Quote or direct sign-up
A catalogue offer is bought online; a specific need gets a costed quote.
3
Design
We design the agent and its guardrails.
4
Integration & testing
We connect your tools to the agent, which is itself hosted in France.
5
Rollout
Going live and training your team.
6
Operation
Continuous supervision and improvement.
Pricing
One package, one agent
A stock management agent (thresholds, lead times, proposals), installed and operated for you.
Agility
Setup + controlled subscription
5,470 € excl. VAT setup
then 474 € excl. VAT/month — you invest at installation and pay a reduced subscription. Ideal for keeping the cost under control over time.
- Installation, configuration and training for your teams
- Operation, human oversight, updates and support
- Sovereign hosting in France, a dedicated and isolated resource
The simplest
Serenity
All inclusive, no setup fee
779 € excl. VAT /month
all inclusive, immediate start. No upfront investment: a single subscription. Ideal for starting quickly and simply.
- Setup included (installation, configuration, training)
- Operation, human oversight, updates and support
- Sovereign hosting in France, managed end to end
100% Sovereign
On site, you own it
8,805 € excl. VAT setup
then 674 € excl. VAT/month · + hardware from 1,058 € (one-off purchase, in addition) — a sovereign computer installed on your premises, maintained remotely. Models run locally, your data returned at the end of the contract. 36-month commitment.
- Hardware installed on your premises (you own it)
- French / European AI models run locally
- Secure remote maintenance (Pro support included)
Not included in the packages: AI consumption (model tokens), re-invoiced at real cost with no margin, and tracked in real time in your client area. Maintenance and supervision subscription for an initial term of 12 months for the Agility package, 24 months for the Serenity package and 36 months for the 100% Sovereign package, renewable; support levels (SLA 72 h / 24 h / 4 h) optional. Bespoke development, additional integrations or exceptional volumes are quoted separately. Support Monday to Friday, 9am to 6pm. Prices exclude VAT.
AI model:
none of the AI models offered currently carries a fixed surcharge. When the selected model carries a cost, that cost is shown when you choose it, before you order, and re-invoiced at the cost incurred, with no mark-up; usage is billed at the publisher's price. Publishers' prices are published in US dollars: the amount re-invoiced is the amount in euros actually borne by Blue Lemon Agent on the publisher's invoice, at that invoice's exchange rate, with no commission or mark-up.
Included components and additional components Components included in the base offer: the Blue Lemon Agent software foundation, the AI models listed in the order journey, the standard channels (Microsoft Teams, Slack, WhatsApp Business, email, website chat, calendars, Microsoft 365 / Google Workspace, file storage, market VoIP telephony, professional social-media pages and accounts, Google Business Profile), hosting in France for the package chosen, backups, supervision, updates and support. If adapting the AI agent to your constraints, your needs or your requests requires other paid components — a third-party publisher's software licence, paid API access to one of your applications, hosting of health data, for which French law requires an HDS-certified host (art. L. 1111-8 of the French Public Health Code), SecNumCloud-qualified hosting, a speech synthesis service, particular hardware —, they are offered to you as an option or on quotation and re-invoiced at the cost incurred; nothing is committed without your written agreement. Where the artificial intelligence model you choose entails an additional cost, that cost is shown to you before you order and re-invoiced to you at the cost incurred, with no margin.
What to expect
Go-live
2 to 3 weeks
Agent designed, channels connected, team trained.
Steady state
4 to 7 weeks
After a few weeks of real use, once the agent's behaviour matches what you expect.
Indicative estimate, adjusted to the options you keep. It is not a delivery commitment.
Our commitment
Four guarantees that matter to your supply
Your stock and your terms stay internalLocal inference or an isolated resource hosted in France; no stock level and no supplier terms entrusted to a third party, no data used to train a model.
Data in France, under French lawYour stock levels and your supplier terms: minimisation and location in France, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
The supply manager keeps the decisionThe agent produces alerts and replenishment proposals, which can be checked and altered; no approval is automated.
Human oversight & traceabilityOn your number of items and suppliers tracked: systematic logging and tracking, in line with the AI Act.
Going further
Related resources
Frequently asked questions
Your questions, our answers
Does the agent place the orders?
No. It prepares the proposal with the quantity, the usual supplier and the context. Placing an order is a purchasing commitment: the decision falls to the supply manager.
Where do the stock levels come from?
From your system, read as it stands. The agent keeps no parallel stock, which avoids any gap between the alert and the reality in the warehouse.
How are the thresholds set?
By you, item by item or by family. The agent applies them as they stand and shows the threshold used with every alert.
Does it take seasonality into account?
It works from your thresholds and your lead times. For anticipation based on history and coming demand, a dedicated forecasting agent complements this use.
Are our supplier terms protected?
Yes. The agent is hosted in France, on local inference or an isolated resource, with the deployment objective of processing and access operated within the European Union and an architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity. Your terms are not used to train a third-party model.
How long does it take to deploy this agent?
A few weeks as a rule, depending on the number of items and the integration with your management system, after a free audit then a phase of design, integration and testing.
Related use cases
Other agents for the supply chain
Let's talk
Let's size up the potential in your stock
15 minutes to frame your items and your thresholds — hosted in France, supervised, with no commitment.