Filter Data component (AI Flowi Workflow)
Filter Data filters a Data object based on a list of keys. The documented interface pairs a Data or JSON source with the string field filter_criteria, then exposes a JSON result.
Filter Data is documented in the catalogue; not production-tested by Ai Flowi.
By Ai Flowi · Published
Catalogue reference
This key-filter reference follows the processing entry in the catalogue for AI Flowi Workflow 1.9.2, exported 9 October 2026: processing.FilterData. Catalogue description: Filters a Data object based on a list of keys.
Browse the Filter Data catalogue entry.
Inputs
Neither of the two exported inputs is marked required. data accepts Data and JSON, while filter_criteria accepts Message; the key-list grammar is not included.
| Name | Type | Required | Accepted input types | Plain explanation |
|---|---|---|---|---|
data | other | No | Data, JSON | The object to filter, supplied by a Data or JSON connection. |
filter_criteria | str | No | Message | String setting for key-based filtering; multi-key separators and key-path syntax are not documented. |
Outputs
| Display name | Name | Type |
|---|---|---|
| Filtered Data | filtered_data | JSON |
Filtered Data is the display name of filtered_data, with JSON as the connection type. The export gives no sample showing which keys remain or what happens to missing keys.
When to use Filter Data
Consider Filter Data when your task concerns keys in one existing object and you want a dedicated filtering step. For example, a synthetic object may contain status alongside a private_note field; inspect the actual result before deciding whether it is suitable for a later consumer.
When to choose another component
Selecting one item from a collection is the job to compare with Select Data, while retrieving one key’s value is covered by Extract Key. For broader JSON operations, follow JSON Operations; for branching on a key comparison, consult Condition.
Example configuration
example configuration, not a recorded test
| Catalogue input name | Example setting |
|---|---|
data | Connect a Data or JSON output containing a synthetic object with status and private_note keys. |
filter_criteria | status |
status is an illustrative single key, not an extra component setting. This one-key configuration avoids assuming a multi-key delimiter; compare the input and Filtered Data in your own run to establish whether the selected key is retained and the other key is omitted.
Limitations
- The description establishes key-based filtering but does not define inclusion versus exclusion, key ordering or a multi-key string format. This page therefore claims no projected output object.
- Missing keys, empty criteria, nested paths and list-valued input behavior are not specified. Optional flags alone do not answer what an empty configuration produces.
- No operator or comparison-value field exists in these two inputs. The component description does not establish filtering collection rows by a condition such as status equals ready.
- A key filter is not documented as validation, authorization or a complete redaction boundary. Confirm the returned object with synthetic extra and nested fields before using it to limit disclosure.
- Table is absent from data’s accepted types. SQL Database’s Result Table needs a separately verified conversion and object shape before this filter can be considered.
Node usage
For budgeting a Filter Data workflow, see the plans and cost guidance.
Related components
Select Data · Extract Key · JSON Operations · Condition · Create Data