How it works
overmust produce a list. The gate runs once per item, with{{item}}and{{index}}available.- Gate it with exactly one of
where,inferordecide. See Conditions and bodies. - With
decide, the result also includes each item’sprobabilities.
When to use it
- Narrow a list before
map, so later steps only see items they can handle. - Prefer
wherewhen a field comparison decides.inferanddecidemake one model call per item; setconcurrencyto run them in parallel.
Input
integer
Maximum
infer verdicts in flight; 1 (default) evaluates one at a time.object
A per-item yes/no question answered by a decision model (only when one is configured); keeps the element when its probability reaches min_confidence. Results add
probabilities, aligned with over.object
string
string
number
0 to 1; default 0.5
string
A decision model role; defaults to decider
string
required
any
The data to judge, usually {{item}}
string
A yes/no question about {{item}}; the element is kept on yes. One judge call per item.
string
Model role for
infer verdicts (any configured role, standard or custom); defaults to the judge role.any
required
The list to partition — usually a template like {{E0.issues}} that resolves to an array.
object
enum<string>
required
Available options:
eq, ne, gt, lt, gte, lte, empty, not_empty, containsany
Comparison operand (omit for empty/not_empty)
any
required
Usually a template about the current element, like {{item.status}}
Result
Later steps read these as{{<step id>.<field>}}.
integer
How many items passed.
object[]
The items that didn’t pass.
integer
How many items didn’t pass.
object[]
The items that passed, in their original order.
number[]
With
decide: each item’s probability, in list order.