executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
core-ocr-and-kieRun docTR end-to-end OCR and KIE inference with predictor APIs, rotation/layout/table options, batching, devices, and output checks.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this sub-skill when the task is to run or debug docTR's Python end-to-end inference pipelines:
doctr.io.DocumentFile or a list of RGB NumPy pages -> doctr.models.ocr_predictor(...) -> Document.doctr.io.DocumentFile or a list of RGB NumPy pages -> doctr.models.kie_predictor(...) -> KIEDocument.Do not use this sub-skill for raw file loading/export format details, model zoo catalogs/custom weights/ONNX export, CLI wrappers, training, datasets, or evaluation scripts. Route those tasks to sibling sub-skills:
doctr-cli and bundled command-line helpers: cli-and-scriptsdet_arch, reco_arch, pretrained, pretrained_backbone, assume_straight_pages, and batch-size choices when reproducibility matters.pretrained=False as an API/shape smoke mode only; randomly initialized detection and recognition models do not produce useful OCR.Document.pages for OCR, KIEDocument.pages[*].predictions for KIE, optional page.layout, optional page.tables, and relative geometries.