Intelligence · Component · office-document-processing
Office/PDF text extraction, scoped storage and document authoring
What it does
Extract bounded text/structure from text-bearing PDF and OOXML, preview DOCX and author/inspect office artifacts. Spreadsheet checks detect missing sheet references and cached errors; they do not calculate formulas.
Core pattern
Bounded document extraction and authoring + previews and structural checks
- Input
- Office/PDF files or structured document content
- Output
- Extracted text, previews, authored documents and validation reports
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| operation | string | extract | extract · preview · author · validate · formula_check · pdf_render · upload_context |
| document_data | — | — | — |
| extract_schema | — | — | — |
| format | string | json | json · csv · markdown · html · pdf · xlsx · pptx · png · svg · text |
| max_pages | integer | 200 | 1–10000 |
Maturity and limits
Native extraction tests plus concrete bundled author/inspect/validate scripts. Phase 2: source paths and implementation inspected statically; tests and external integrations were not executed. Maturity rates repository implementation, not a verified production deployment.
Scanned PDFs need a separate OCR engine; these PDF text extractors do not perform OCR. Static formula checks do not recalculate Excel formulas. ZIP expansion and document limits need resource controls.
2 sentences citing source paths or vendor names omitted for publication.
Used in pipelines
Industry tags
- Legal servicesTransferable use
For case-file organization and source-linked summaries, use bounded document extraction and authoring followed by previews and structural checks.
- HealthcareTransferable use
Support “Confirm and archive notes” with bounded document extraction and authoring followed by previews and structural checks.
Skills pack
Contract scaffold and deterministic mock
Download · 14KBSHA-256 d0ff4f0a5b19b5910a2105a43c6a5a74fb1b94fc6c1bca2a9da4e8260dab03f5
Related components
-
Multi-image OTA rate extraction
Paired screenshot field recognition + structured comparison records
Working 1 pipeline -
Disclosure collection and text extraction
Document index collection + cleaned text and normalized references
Working 1 pipeline -
Legacy data and provenance migration
Legacy record mapping and import + orphan and completeness checks
Working 1 pipeline -
Podcast and video catalog ingestion
Media metadata normalization + visibility and retention rules
Working 1 pipeline