Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 5 additions & 7 deletions skills/uipath-maestro-flow/references/document-pipeline.md
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ Signatures: `documentClassify({ fileRef, pageRange?, splitPages?, modelConfig? }
model: { modelName: 'invoiceixp-cef0d447-ixp',
folderKey: 'c4359cde-55f0-4f0e-9322-c6cdce74ab4c' },
}))
.step('total', script({ code: 'return $vars.extract.output.result.output;' })) // warns until the SDK typing catches up
.step('total', script({ code: 'return $vars.extract.output.result.output;' }))
```

## At a glance
Expand Down Expand Up @@ -80,12 +80,10 @@ Classification lands at `out('classify', ...)`. What the node emits at run time
is a processed document, with the extracted values at `result.output`, one
property per taxonomy field, and their page provenance under
`result.attribution` ([Reading what it extracts](extract-document-node.md#reading-what-it-extracts)).
The two routes embed different typings for it, though: `nodes create` embeds
that current shape, while the SDK factory still embeds the older
`ExtractionResult` shape — so on a factory-built flow, a read of
`result.output` validates with an `EXPRESSION_DIAGNOSTIC` warning until the SDK
catches up. `ExtractionResult` as a runtime shape belongs to the
published-model node ([ixp.md](ixp.md)), not this one.
Read a value with `out('extract', 'result.output.<field>')`. `check` refuses a
read of a key the document does not have (`EXTRACT_READ_UNKNOWN_FIELD`), and
names the spelling to use. `ExtractionResult` is one of those keys: it is the
published-model node's shape ([ixp.md](ixp.md)), not this one.

## Evidence boundary

Expand Down
Loading