Flow guide

Graxeon Flow is the ingestion and automation pillar: connect an external source, decide what starts it, map its fields onto a collection's schema, and Flow writes the records in as typed nodes — no code required.

What is Flow

Flow is built from three pieces that combine into a pipeline:

• Connector — a saved, credentialed configuration of a source. Reusable across multiple Flows.
• Flow — a Connector plus a field mapping, bound to one collection and node type. This is the unit you build and turn on or off.
• Trigger — what starts a Flow: a schedule, an inbound webhook, or a manual run. A Flow can have any number of Triggers, or none at all.

Records that come in are matched against the collection's schema and written in as nodes through the same validation path Studio uses — so data that arrives through Flow is held to the same structure as data entered by hand.

Connectors

Flow ships with four connector kinds:

• Webhook (inbound) — accepts a JSON POST (a single object or a list of objects) whenever the sender pushes to it.
• HTTP poll — a scheduled GET against a JSON REST endpoint, with an optional bearer token and a dotted path to the record list within the response.
• RSS / Atom feed — polls a feed and turns each new item into a record.
• CSV upload — a CSV file you upload from a Flow's own page; each row becomes a record.

A Connector is created once and can back several Flows — for example, one "Stripe events" webhook connector feeding both a Customers Flow and an Invoices Flow, each with its own mapping into a different collection.

Creating a Flow

Building a Flow takes four choices:

• Connector — which source to pull records from.
• Collection and node type — where the records should land, and which schema type they should become.
• Field mapping — for each field you care about, which source field maps to which target schema field.
• Dedupe field (optional) — a target field carrying the source's own identity. When set, a record whose mapped value already matches an existing node updates that node instead of creating a duplicate — essential for anything that polls or pushes repeatedly.

A new Flow has no Trigger yet, so nothing runs automatically until you add one — see below. Until then (or in addition), it can always be run by hand.

Triggers

Triggers are first-class and independent of the Flow they belong to — a single Flow can have a schedule Trigger and a webhook Trigger, several webhook Triggers from different callers, or none. Every Trigger has its own enabled/disabled switch, so you can pause one path into a Flow without touching the others.

• Schedule — polls the Flow's connector on an interval you choose. Only available for pollable connectors (HTTP poll, RSS).
• Webhook — gives you a unique URL; a POST to it runs the Flow immediately with the request body as the record(s). The token in the URL is the only authentication, so treat it like a secret.

Running manually

From a Flow's own page, pollable connectors get a Run now button that fetches and processes immediately; CSV connectors get an upload form. Both use the same execution path as a Trigger firing, so results show up in the same run history either way.

Dashboard

Flow's home page gives you an at-a-glance view across every Flow:

• A hero chart of runs per day, stacked by outcome, over the last two weeks
• Upcoming — the next scheduled Triggers due to fire
• Actions — the most recent runs across every Flow, regardless of outcome
• Issues — the most recent runs that failed or only partially succeeded

Plans

Automation is a Pro and Team feature. Free accounts can browse the Flow app but creating or enabling Connectors, Flows, or Triggers requires an active Pro or Team subscription — see pricing. Self-hosted Graxeon Core installs with no Stripe keys configured run Flow with no billing gate at all.