Cleaners

A Cleaner is a set of rules that runs over the rows a flow has already extracted. It standardises formats, converts currencies and units, translates text, computes new columns, looks values up against your own data, and can trigger actions when a row breaks a rule.

Extraction answers “what does this document say?”. A Cleaner answers “what shape does that have to be in before it can go into our systems?” — one date format, one currency, part numbers matched to your catalogue, a total that adds up.

A Cleaner is built base fields first. Base fields are the columns coming in; everything else is computed on top of them. You can define them by hand, import them from a flow, or parse them from a spreadsheet header row.

1. 1Go to Cleaners and create a new one. Name it after the data, not the document — Invoice line items rather than Cleaner 3.
2. 2Choose a data source for the base fields: Manual to define them from scratch, From Flow to import an existing flow's output fields, or From File to read them from a CSV or Excel header row.
3. 3Add computed fields — the enrichment types in the table below. This step is optional; a Cleaner with only base fields is a valid pass-through.
4. 4Configure the sweep output: an email address, a webhook URL, or neither. Also optional.
5. 5Link the Cleaner to a flow so every new run is swept automatically, or leave it standalone and run it against uploaded datasets.

Every column in a Cleaner has a type that decides what it does. Base fields pass values through; the other types compute a value from the row, from your Buckets, or from an AI judgement. This is the complete set.

A field can be excluded from the output while still being available to computed columns. That is how you use an intermediate value — a raw amount, a lookup key — without shipping it downstream.

A Conditional Actions field is a rule: when a row matches these conditions, do these things. It is the only field type that changes what happens to the run rather than what a cell contains, and it is how validation failures become something other than a number in a table.

Conditions are grouped, and groups combine with AND or OR, so you can express things like (total > 10,000 OR currency is not USD) AND supplier is not on the approved list. A rule with no conditions at all fires on every row.

Worked example. On an invoice Cleaner, add a Conditional Actions field with two rules. The first matches rows where the line total does not equal quantity × unit price and sends the run for review. The second matches any invoice over your approval threshold and emails the finance lead. Everything else flows straight through to the Bucket without a human touching it.

A Calculated field evaluates an arithmetic expression against the row it is on. Reference other columns by name in braces, and start the formula with an equals sign if you like — both ={Quantity} * {Unit Price} and {Quantity} * {Unit Price} work.

Summary fields aggregate across every row in the sweep instead of computing per row. They answer “what is the total of this document?” without you having to sum the rows downstream.

The HS Code field assigns a customs tariff code to each row from its product description. It works down the Panama tariff schedule (HS 2022) — section, then chapter, then heading, then national tariff line — applying the chapter legal notes rather than pattern-matching a description against a code list.

• Pick which columns the classifier reads — usually the product description, sometimes with material or use alongside it. Leave it unset and the whole row is used.
• Add instructions for the cases your catalogue gets wrong: how to treat kits, spare parts, or goods that could sit in two chapters.
• Rows are classified independently, so one hard row does not affect the others.

Between the computed columns and delivery, a Cleaner can drop junk rows, reshape long data into wide, choose which columns ship, and send the result to an email address or a webhook.

A sweep is one execution of a Cleaner over a batch of rows. Sweeps run automatically when a linked flow finishes, or on demand when you upload a dataset to clean. Each sweep is billed on the number of non-empty cells it processed, not the number of documents.

Cleaning credits are charged per 500 non-empty cells, rounded up, with a one-credit minimum. Empty cells are not counted, so a wide table with many optional columns costs less than its dimensions suggest. Every sweep records the cells processed and the credits it used.

Open a sweep from the Cleaner's Sweep History to see the rows it produced, the credits it consumed, and per-row errors — a formula that divided by zero, a lookup that matched nothing, a value that failed validation. Errors are attached to the row that caused them, so a single bad row does not fail the sweep.

Cleaning happens after extraction and before delivery. That order matters: everything downstream — the review screen, the webhook payload, the Bucket rows, an agent's input variables — sees the cleaned output, not the raw extraction.

1. 1The flow extracts rows from the document.
2. 2The linked Cleaner sweeps those rows: formats, conversions, lookups, computed columns.
3. 3Conditional Actions fire — rows are dropped, notifications go out, review may be requested.
4. 4If review was triggered, the run pauses and a reviewer sees the cleaned table.
5. 5On completion, results are delivered: email, webhook, Bucket, form fill, or a chained agent.

The Pipeline Map shows this for your own workspace, including which flows share a Cleaner.