One layer under every agent and every person.

LeafCraft is the Hybrid Intelligent Enterprise platform. People and AI, one team, under your rules, with your memory and your budget. It runs on your own servers or cloud.

Every hand-off between an agent, a person or a system goes through one layer that applies your rules, asks the right person and writes down what happened. That's why your rules hold on every run, not just the ones someone checks.

One teamYour rulesYour memoryYour budget
Product screen · example dataLeafCraft console
Trusted by
MiVi
mediamint
Bhashini
Neurasix
Wyra

How a run works

Nothing moves without going through your rules.

Five steps, the same on every run.

Work arrivesLeafCraft · your rulesClaims agentA person signs offRecordedClaim C-20418North region · webhookFacts firstScheme S-118 · INV-7731“Approve.”Matches scheme and invoiceAbove the agent's limitSo a person decidesPriya Nair approvesIn Slack · 14:12Who · why · cost3 h 31 m · 1.84 of 4.00One run · example data
  1. 01Work arrives
    WebhookKafkaSlack
    Claim C-20418North region · 10:41:01

    API, webhook, schedule, message or queue

  2. 02Facts first
    • Scheme S-118
    • Invoice INV-7731
    • Dispatch quantities

    Your code pulls what the task needs

  3. 03A role acts
    Claims agent"Approve. Matches the scheme and the invoice."

    Agent or person, within its limits

  4. 04Rules decide
    Above the agent's limit
    Next agentA person signs offStop

    Next agent, sign-off, escalation or stop

  5. 05Recorded
    Who
    Priya Nair
    Why
    Scheme confirmed with the regional team.
    How long
    3 h 31 m
    Cost
    1.84 of 4.00

    Who, what, why, how long, what it cost

One run · example data

The console

See every agency, its spend and its errors. Stop any of them.

The Control page shows cost, sessions, errors and plan usage across every agency.

  • Spend, liveLLM spend per agency, broken down by agent, model and run.
  • Errors that stop safelyA step that keeps failing is stopped and handed to a person, instead of looping.
Product screen · example dataLeafCraft console

Getting started

Start from a working team, not a blank page.

01

Pick a template

20 ready-made agent teams, built the way real departments work, with sign-offs and an audit agent already in place.

02

Shape it in Studio

Change roles, rules, sign-offs and budgets. Try a single agent in the Playground before it touches real work.

03

Run it where you want

One command puts it in your AWS, Azure or Google Cloud account. Or run it in your own Docker or Kubernetes setup.

Read the docs
SalesProposalsQuote-to-cashMarketing opsCustomer successSupportEngineeringCloud costFinanceProcurementProduction planningQualityMaintenanceWarehouseServices deliveryRecruitingBench managementStarter

Questions teams ask

Is LeafCraft an agent builder or a governance layer?

Both, on purpose. You can build agents in LeafCraft, and you can bring agents from 40+ other platforms. Either way they run under the same rules and leave the same record. Agents you build or bring belong to you.

Does it replace our existing tools?

No. Agents read and write to the systems you already use, and people get requests in Slack, Teams, WhatsApp (coming) or email. Nothing gets rebuilt.

Where does it run?

In your cloud account, your VPC or on-prem. Run records live in your own Redis or Postgres.

Can a model decide to skip a step?

No. Routes are plain conditions checked by code the same way every time. A model can't overrule them.

Discover every agent

You can't govern what you can't see.

LeafCraft connects to your ecosystem and finds every agent you run: the ones vendors shipped inside your software and the ones your teams built. Then each one gets an owner, comes under your rules, or is removed.

Shadow AI endsEvery agent ownedUnwanted agents removedMemory starts from day one
How discovery works

Show us the process you can't afford to get wrong.

We'll map it onto LeafCraft with you: who does what, where a person signs off, what it costs and what gets recorded.