AI Copilot — plain-language operator for your platform

Not a chatbot. A grounded operator that calls the exact same Commands and Queries your engineers wrote — with full audit trail, EU-sovereign providers, and hard cost caps.

AI-native, the way regulated tenants actually need

Talk to your platform. The platform actually does the work.

AI Copilot is the seventh CleverInit module — and the platform's first AI-native surface. Install it on your tenant and your admins can talk to the platform in plain English, Dutch, or German. Create draft invoices. Find customers who haven't paid. Summarize a week of chat. Diagnose why a subscription went past-due. The Copilot does it by calling the exact same Commands and Queries your engineers wrote — never by guessing, never by hallucinating, always with full audit trail.

What makes it different: it is not ChatGPT in an iframe. The LLM is the orchestration brain; CleverInit's CQRS handlers are the hands. Every consequential answer is backed by a tool invocation whose inputs, outputs, latency, and cost are recorded. Every mutating action requires explicit user confirmation. Per-tenant cost caps prevent runaway loops. Per-tenant provider selection (OpenAI, Anthropic, Azure OpenAI, Mistral, Ollama) gives you the EU sovereignty story your customers will ask about on the first compliance call.

  • Installs as a marketplace module — federated pane, topbar button, chat-bot bridge
  • Streams responses token-by-token over SignalR — same hub pattern as the chat module
  • Every tool invocation is logged with redacted arguments, result hash, cost, and latency
The wow moment

A tenant admin types "Create a draft invoice for Acme Holdings for €5,000 due in 14 days". The Copilot picks the Invoicing tool pack. Permissions check runs. FluentValidation runs. The same MediatR pipeline your engineers use runs. A draft invoice appears in the panel with Acme pre-linked. The admin clicks "Issue". Done. No new endpoints. No custom integration. Every module on your tenant just became conversational the moment Copilot read its tool pack.

What it can do

Real commands. Real outcomes. Real audit trail.

Every Copilot answer is grounded in a tool call — a typed Command or Query that already exists in your platform. The LLM picks the right one; your CQRS handlers do the work.

INVOICING
YOU SAY
>Create a draft invoice for Acme Holdings for €5,000 due in 14 days.
Copilot calls CreateInvoiceCommand. Validator runs. Permission check runs. Draft invoice appears with Acme pre-linked, awaiting your one-click Issue.
CUSTOMERS
YOU SAY
>Show me the customers who haven't paid in 30 days.
Copilot queries the Customers + Invoicing tool packs, joins the projection in-platform, returns a paginated list with last-paid date and outstanding amount.
CHAT
YOU SAY
>Summarize this week's chat in #support.
Copilot reads only conversations you participate in, summarizes themes, surfaces unanswered threads, and posts the digest back to you privately.
SUBSCRIPTION
YOU SAY
>Why did the subscription for customer 412 just go past-due?
Copilot walks the SubscriptionRenewedV1 and PaymentFailedV1 event chain, identifies the failed renewal cause, and links you to the customer dossier.
PAYMENTS
YOU SAY
>List all payments above €2,000 from the past week.
Read-only at v1 — Copilot will list and filter payments, but never initiate, refund, or void a charge without your explicit approval.
NOTIFICATIONS
YOU SAY
>Notify Sarah that the Q4 report is ready and CC the finance group.
Copilot drafts the notification, shows you the recipient list and message, waits for your confirm, then publishes via the notifications module.
Grounded tool-use, not raw chat

How the Copilot actually answers

Most AI products bolt an LLM onto a REST monolith and let the model guess the endpoints. CleverInit is different: every state change is already a Command, every read is already a Query. The Copilot just calls them.

01

Ask in plain language

You type or speak. Dutch, English, German — auto-detected. Voice is delivered via the same SignalR hub your chat module uses.

02

Copilot picks the right tool

From a registry every installed module contributes to. Each tool has a typed input schema, a typed output, a required permission, and a scope.

03

Permission + validation, twice

Your effective permissions are evaluated before the model sees the tool. FluentValidation runs on the input. Defense-in-depth, identical to your hand-written endpoints.

04

Tool runs through MediatR

Same pipeline as every other Command — Logging, Validation, Cache Invalidation, Transaction. The Copilot adds nothing custom; it reuses the platform you already trust.

05

Audit + stream back

Result hash, cost, latency, redacted args go to an append-only audit table. The answer streams back to you token-by-token, with tool-call status updates.

This is not ChatGPT in an iframe. The LLM is the orchestration brain; your CQRS handlers are the hands. Every answer is backed by a tool invocation whose inputs and outputs are recorded — the only pattern that gives a Dutch fintech or German healthcare tenant an auditable trace they can defend on a compliance call.
Per-tenant provider selection

Pick your own AI provider. Per tenant. Per workload.

A Dutch healthcare tenant under GDPR can pick Mistral or Ollama hosted in the EU. A US fintech can pick OpenAI or Anthropic. Credentials are encrypted via Data Protection — same pattern as Stripe and Mollie keys today.

OpenAI

GPT-4 family, broad ecosystem

Most capable general model family. Familiar tool-use envelope. The default choice for tenants who already standardize on OpenAI.

Anthropic (Claude)

Premium reasoning, EU inference

Strongest performance on long-context, multi-step tool chains. EU-region inference now widely available. Recommended default for new tenants.

Azure OpenAI

Microsoft-hosted, EU agreements
EU DATA RESIDENCY

Same wire protocol as OpenAI behind an Azure-managed endpoint. The right answer for tenants who already sign EU data-residency agreements with Microsoft.

Mistral

EU-headquartered, sovereignty
EU SOVEREIGN

Headquartered and operated in the EU. The sovereignty answer for healthcare, government, and fintech tenants whose compliance team will ask the question.

Ollama (self-hosted)

On-prem, zero egress
ON-PREMISES

Tenant-hosted on-prem inference. No external network egress. Lowest cost, highest privacy posture. Configure once per tenant; the rest is identical.

EU data residency, sovereignty for regulated tenants, and tenant-level cost optimization — all from a single provider selector. No custom infrastructure on your side.
Every module becomes conversational

One tool registry. Every installed module ships tools to it.

Each CleverInit module exposes its Commands and Queries as a tool pack. Install the module — its tools light up in Copilot. Uninstall it — they disappear. No new endpoints, no custom integration, no model retraining.

Customers

  • Find a customer by name, email, or VAT id
  • Create a new customer with a starter note
  • Archive or restore a customer record

Invoicing

  • List invoices by status, date range, or customer
  • Draft an invoice with line items and due date
  • Issue an invoice (requires confirmation), record manual payment, void

Payments

  • List payments above an amount in a date range
  • Query the status of a specific payment by id
  • Read-only at v1 — no charges, refunds, or voids from the Copilot

Subscription Manager

  • Show a customer's current subscription and renewal date
  • Draft a plan change effective at next renewal
  • List past-due subscriptions across the tenant

Chat

  • Summarize a conversation you participate in
  • Post a message on your behalf into a thread you are in
  • List unread mentions across your channels

Notifications

  • Send a notification to a user or group
  • List a user's recent notifications by channel
  • Mark a notification as read or archived

Identity (read-only at v1)

  • Find a user by email or display name
  • List a user's effective roles in this tenant
  • List active permissions — no mutations from the Copilot at v1
FOR SOFTWARE COMPANIES & ISVs

Implement ICopilotToolProvider once. Your module is now conversational.

Software houses and ISVs building modules on top of CleverInit get the biggest win. Implement the ICopilotToolProvider contract in your module's Contracts package, declare your tools, and every tenant who installs both your module and AI Copilot can drive your features by voice or text. No language model expertise required. No prompt engineering. The LLM handles the language; your CQRS handlers do the work.

Safety posture

Designed for tenants whose compliance team will read this page.

The Copilot ships with the strictest defaults that still feel useful. Tenants can loosen each one — but the defaults are designed to survive a CISO review without changes.

Permission check, twice

Every tool is gated by a [HasPermission] string on both the controller boundary and the handler. A user without the permission cannot even see the tool in the model's request envelope.

Confirmation for every mutation

Mutating tools (draft, issue, send, post) require an explicit click before they run. Read-only tools execute inline. Auto-execute for safe reads is gated by a per-user permission, opt-in.

No payments, no identity mutations

v1 explicitly forbids the Copilot from initiating charges, refunds, mandate creations, password resets, role changes, or user creations. Read-only at most for the regulated surfaces.

Provider credentials encrypted

Every API key is encrypted at rest via IDataProtectionService — the same pattern used for Stripe, Mollie, and PayPal keys today. Plaintext is held only inside a request scope, decrypted at the call site, never logged.

Hard cost caps default ON

Per-tenant daily and monthly caps, per-user daily caps. Breaches publish CopilotCostLimitBreachedV1 and suspend the session with a clear message. A runaway loop is bounded in cost, not in apology.

GDPR Article 17 hard delete

Sessions can be archived (default) or hard-deleted. Hard delete is gated by a separate AiCopilot.Erase permission — because not every operator should be allowed to destroy audit evidence, even their own.

Install AI Copilot. Talk to your platform.

Install AI Copilot from your marketplace, point it at a provider (OpenAI, Anthropic, Azure, Mistral, or your own Ollama), and your tenant admins can drive every installed module in plain language by this afternoon.

Install AI CopilotTalk to engineering

Build smarter. Launch faster. Scale effortlessly.

Tell us where you want to take your business. We'll show you the fastest path to get there — whether you want a fully managed platform, a custom module built exclusively for your tenant, or a fully branded product to sell to your own clients.