Data & Variables¶
Data moves through a flow as it runs. A run starts with something - a form submission, an API payload, a record - and each step adds to what the run knows, so later steps can act on what earlier ones produced. Getting a value from where it is to where it is needed, in the shape the next step expects, is what this section is about.
It is one part of building a flow. Deciding the path a run takes is Flow Control; giving an agent its instructions and the tools it can reach for is AI in Flows; calling other systems and handing work to other flows is Integrations & I/O.
Find the page below that matches what you are trying to do.
Use one step's output in a later step¶
The everyday case: a block ran, it produced something, and a step further down needs it. You point the later step at the earlier one's result and pick out the piece you want. See Passing Data Between Blocks.
Keep a value for later in the run¶
When a value has to outlive the block that produced it - a running total, a flag you set early and check late, a message you assemble in pieces - you hold it in a named variable that the rest of the run can read and update. See Holding Values in Variables.
Change data into the shape a step needs¶
Data does not always arrive in the shape the next step wants. A date needs reformatting, a list needs filtering or sorting, a set of numbers needs totalling, two values need merging into one object. See Reshaping Data.
Carry data beyond a single run¶
Every run starts fresh, so a value written in one run is gone by the next. When state has to survive - a running count across runs, a cursor marking how far you got, a token to reuse - or be shared with another flow, you need a store that outlives the run. See Sharing Data Across Runs & Flows.
Related, outside Data & Variables¶
- Variables & Data Buckets - what a variable is, and how a Data Bucket groups them.
- Expression Editor - the dialog where you build a value from references, operators, and fixed text.
- Blocks - how a block's result becomes readable by the steps after it.
- Shared Memory - the flow's own store, kept between runs.