PRODUCTS · AKILI · LICIA
LEYSCO AKILI · LICIA — EARLY ACCESS

The assistant is the architecture.

Leysco Licia is an ERP AI assistant that queries data and drafts actions through the same typed services as the screen. Every action is entitlement-checked and audited. Early access.

2026-06-01first boot
21modules designed
30sprints closed
Governed AI · human approval
ON LEYSCO AKILIEarly accessEarly access. Shape the platform as a design partner.
The category

AI-native ERP assistant / governed agent platform

The names you'll be comparing

SAP Joule · Microsoft Copilot (D365) · Salesforce Agentforce

The difference

They bolt an assistant onto a finished suite. Here the assistant and the screen call the same governed services, under one permission model for humans and AI.

agent → tools/listSCOPED BY ROLE
semantic.query
propose_journal_entry → DRAFT
resolve_master_data
ingest_document
… typed tools
ONE GOVERNED SURFACE — FOR LICIA AND ANY EXTERNAL AGENT

Every AI action is a governed tool call.

The gateway

An MCP tool server exposes the platform as typed, permissioned tools — to Licia and to any external agent. The propose-to-draft pattern: propose_journal_entry returns a DRAFT. The agent proposes; the platform decides.

JOURNAL DRAFT CHECKSLEYSCO AKILI
With journal checks switched on, re-read each journal draft after it is written and confirm its total and that debits equal credits.
Open a draft's evidence record: the checks it passed, who asked for it, what changed and its place in the audit chain.
IF THE TOTALS DON'T READ BACK, THE DRAFT STOPS FOR REVIEW

Inspect the checks behind the draft.

Journal draft checks

Switch on journal checks, and each journal draft from Licia is checked before it is written, re-read afterwards and confirmed to balance. If the totals do not agree, it stops for a person to review.

action 0#a91f… action 1#3ce7… action 2#7b02… action 3#d144… action 4#e8aa… EACH ACTION HASHES THE PREVIOUS — TAMPER-EVIDENT · APPEND-ONLY

Audited like a ledger, because it writes to one.

Trust plane

PII routing, injection defense, output-integrity checks, cost ceilings — and a hash-chain audit of every AI action. The same discipline the ledger gets, applied to the model.

MODEL ROUTINGPROVIDER-AGNOSTIC
ProviderRoutingTier
Claudetiered by taskcloud
Ollamaself-host default$0 local
MODEL CHOICE IS A CONFIGURATION ROW, NOT A PROCUREMENT CYCLE

Your ledger never has to leave the building.

Model freedom

Claude tiers routed by task — or local Ollama as a $0 self-host default. Provider-agnostic by architecture, so model choice is a configuration row, not a procurement cycle.

ANSWERS FROM YOUR DOCUMENTSLEYSCO AKILI
Search only your company's published documents, keeping internal documents to the people permitted to see them.
Quarantine documents that carry hidden instructions aimed at the AI, and drop any such passage before it reaches an answer.
Reject any answer that cites a passage Licia did not retrieve.
Return a clear no-answer response when retrieval or citation checks fail.
PUBLISHED DOCUMENTS · PERMITTED READERS · RETRIEVED SOURCES

Answers with their sources attached.

Governed knowledge

Licia answers from your own published documents and shows the source behind every answer it gives from them.

The platform under Licia.

In this line
Document ingestion (email/WhatsApp/PDF → governed draft)Early accessReceipt in, postable draft out.
Master Data ResolverEarly accessMoney-safe fuzzy entity resolution.
Semantic Layer — governed metricsEarly accessOne measure, three renderings.
MCP Tool Server & agent gatewayEarly accessThe governed surface itself.
Schema engine (~375 schemas → database)Early accessConfigure-not-code substrate.
Authorization — one chain, humans + AIEarly accessRBAC/RLS/FLS/SoD incl. 12 ai.* permissions.
3-tier multi-tenancyEarly accessTenant-bound isolation and secrets.
Licia standalone — bring your own ERPEarly accessSame assistant against SAP B1, ByD, BC, NetSuite.
AI eval harness & injection-defense toolingDesignedQuality gates for AI features.
Answers from your documentsEarly accessLicia's sourced answers from your published documents.

Watch an agent get governed.

Signature · Early access · Sample data

A deterministic replay of an MCP session: the agent lists its tools, asks a governed question, proposes a journal entry — and the platform answers with a gate, not an execution.

mcp session · replay · 2026-06 dev env
  1. agent→ tools/list
  2. akili← typed tools · scoped by role: semantic.query, propose_journal_entry, …
  3. agent→ semantic.query { measure: "expenses.total", period: "2026-05", by: "cost_center" }
  4. akili← 12 rows · governed model · RLS applied for this principal
  5. agent→ propose_journal_entry { debit: "6200-FUEL", credit: "1000-PETTYCASH", amount: 18500 }
  6. DOPSGATEjournal checks passed · balance asserted · read back OK → returned as DRAFT je-2026-000418, not posted
  7. akili← audit: action hash-chained · cost metered · awaiting human approval
THE AGENT PROPOSES. THE PLATFORM DECIDES. · EARLY ACCESS
1.66M+invoices

The team behind Leysco Akili runs 1.66M+ posted invoices on Leysco100, its sibling house. Licia comes with that team and its domain depth.

Two independent, co-equal houses. Choose by fit. The native immutable multi-book general ledger is designed.

Built for operations that…
  • want to shape an agent-operated ERP from the inside, as a design partner
  • need AI that can act — under entitlements, gates and audit — not just chat
  • have a data-residency posture that prefers a self-host model tier
  • want to move early on AI-native ERP
The old way → this way

A chat box beside the ERP, guessing at screens → a governed caller inside it

Choose by fit. Need it running this quarter, on Kenyan rails — Leysco100. Building the agent-operated enterprise and want to shape it from the inside — Leysco Akili, in early access.Both come with the same team.

Shape the assistant from the inside.

Early access. Nairobi · EAT.
Request early access
Talk to us · WhatsApp +254 780 457 591