What is Pipory
Pipory is an AI-native visual workflow automation builder - design node-based workflows on a canvas and let them run that part of your business.
Pipory is an AI-native visual workflow automation builder, in the same family as n8n or Make. You design workflows as a graph of nodes on a canvas, connect them with edges, and run them through a durable execution engine - with every node's live status streaming back to the canvas while it runs.
The difference from most tools in this space is the pricing model: there is no per-task or per-step metering. Cost does not grow with how many nodes fire or how many workflows you run, and durable waits (delays, schedules) cost nothing extra.
The three pieces
- Canvas. A React Flow-based editor where you drag nodes from a palette, wire them together, and configure each one in a dialog. See Workflows.
- Engine. A durable execution engine (built on Inngest) that walks the graph in topological order, resolves Handlebars templates against accumulated context, and persists a detailed log of what every node did. See Executions.
- Credentials. Every third-party API key or token you connect - Slack, Anthropic, Postgres, Stripe, and dozens more - is stored per-user and encrypted at rest. See Credentials.
Quick tour
A minimal workflow looks like this:
- A trigger node starts the run - manually from the editor, on a webhook POST, on a cron schedule, from a linked Google Form, or on a Stripe event.
- One or more action nodes do the work - call an HTTP API, post to Slack or Discord, run an LLM prompt, query a database, or call one of the 200+ dedicated app nodes.
- Flow control nodes shape how the run moves - branch on a condition, loop over an array, delay, or route to an error handler.
Every node reads from the accumulated run context and writes back into it, so later nodes can reference earlier outputs with a Handlebars expression like {{httpRequest.body.email}}. See Workflows for exactly how that resolves.
Where to go next
- New to Pipory: start with Getting started and build a webhook-to-Slack workflow end to end.
- Want the mental model first: read Workflows, Triggers, and Flow control.
- Looking for a specific node: jump straight to the node reference.