Focus on your business.
Zervanor drives your automation engine.
Workflows, connectors and agents your product runs on — with local invoicing, payment and logistics rules already modelled. Below is what the engine runs today.
Every entry is grounded in a versioned connector pack the engine can actually execute — a trigger, its steps, and the agents that handle the judgement.
Prefer to build it yourself? A visual canvas with real depth.
Chat-to-Build and the visual builder produce the same thing — one real workflow. Snap triggers, control flow, and actions together as blocks your team inspects, governs, and extends. Everything the chat assembles, you can open and edit here.
Visual workflow builder
Compose flows on a canvas, inspect every run, and open any Chat-to-Build draft to refine it by hand.
43 node types
Triggers, control flow (switch, loop, parallel, filter, merge, wait), data transforms, connector actions, agent steps, human approval, code, and Tables — the platform’s own data store, so an automation can remember between runs.
Agent hosting included
Long-running AI agents run as isolated pods on the same runtime — no extra infrastructure.
Real workflow blocks in the Zervanor builder.
The building blocks
The real node palette — the same step types the chat grounds against and the engine runs.
cloudeventSchedule / CronscheduleWebhookwebhookManualmanual conditionSwitch (multi-way)switchLoop (for-each)loopParallelparallelFilterfilterMergemergeDelaydelayWait (durable)waitStopterminateNo-opnoop edit_fieldsTransform (JSONata)transformSortsortSummarizesummarizeAggregateaggregateTable (org data store)table_action connector_actionRun agentagent_actionAgent taskagent_taskHuman approval (HITL)human_taskCode (Deno sandbox)script_actionSub-workflowworkflow_actionEmit eventemit_event A catalog that compounds — and doubles as agent tools.
Every connector is a versioned pack validated against a canonical schema. Certify one once and it is instantly available to both your workflows and your AI agents.
Live counts from the platform — plus thousands more providers mapped and queued in the pipeline.
Run on the platform — or on your own device.
The same agent runs in the managed cloud or on your own machine, draws on a routed mix of cloud and on-device models, and is reachable through a chat interface.
Cloud or your device
The same agent — one manifest, one SDK — runs in a managed cloud pod or on your machine via the Zervanor Desktop app. It connects outbound, so it works behind a firewall.
Smart model routing
One catalog of cloud, platform and on-device models behind named routers with automatic fallback. Choose a model per agent; run inference locally when you want.
Conversations
A persistent, model-routed chat that can drive your connectors and agents as tools — with a confirmation step before anything changes.
Desktop runtime host today; a mobile host is on the roadmap.
Enterprise-grade underneath, SMB-simple on top.
Multi-tenant isolation, access control and observability from day one — the operational floor a product has to stand on before it ships to anyone.
Tenant isolation
Per-tenant Kubernetes namespaces keep every organization separated — from day one, not an upsell.
Access control
Role-aware access and identity integration for teams that need operational control.
Observability
Event history, OpenTelemetry tracing, and a token-free public status page.
Start building your first automation
Create a free account and build on the live platform — describe what you want in chat, or compose it on the visual canvas. No credit card required.