Split a parsed document into named sections using an LLM. Consumes ai.parse output; emits per-section page ranges and text ready for downstream ai.extract via control.parallel_map.
ai.split, Split a parsed document into named sections using an LLM. Consumes ai.parse output; emits per-section page ranges and text ready for downstream ai.extract via control.parallel_map.
What this section looks like in the document. Use the document’s own terminology (e.g. “Príloha 2 / Anlage 2”) and visual cues. Multilingual variants belong here.
Natural-language hints fed to the LLM about what marks the END of this section (e.g. [“PRÍPAD PORUŠENIA ZMLUVY”, “Koniec prílohy”]). The LLM uses these as guidance, not a regex match, so casing variants, missing diacritics, and multilingual phrasings all close the section correctly.
Override the estimated per-window token ceiling. Oversized pages are subdivided with overlap while retaining their original page index. Defaults to env SPLIT_WINDOW_TOKEN_BUDGET or 20000. Inputs requiring more than 64 LLM windows are rejected before calls begin.
Sections found in the document, in page order. Absent sections are omitted.
Show splits properties
Section name from the config
[startIndex, endIndex] inclusive, 0-based page indices
LLM confidence at the anchor page. Coarse enum (low | medium | high), numeric scores cluster meaninglessly at 0.85-0.95.
LLM’s justification for the anchor, useful for debugging
Structured start-anchor evidence, parse this instead of regexing over notes.
Show evidence properties
Verbatim heading text the LLM cited as the section start
Page where start_heading_text appears
Set when the LLM detected an end-of-section cue (matched against the section’s endHints or an explicit closing marker). Absent when the section was closed by continuity-fill alone.
Show end_evidence properties
LAST page of the section (inclusive)
LLM’s justification for the end. Pair with start notes for reconciliation.
Joined per-page content for this section, ready for downstream ai.extract
Raw per-page records covered by this split (in page order). Iterate when downstream needs per-page context (e.g. control.parallel_map over section pages).
Show pages properties
Whether this page is direct LLM evidence (anchored) or continuity-filled (inferred)
Pages with direct LLM evidence
Pages assigned by deterministic continuity fill
The same sections keyed by config name, so a downstream step can reference one directly: {{ steps.<split>.output.sections.<name>.page_range }}. Prefer this over filtering splits. On a duplicate name the last wins.
