STOP. READ THIS ENTIRE SKILL.MD BEFORE CALLING ANY ENDPOINT.

learn-datalake

learn-datalake is the user-facing orchestrator for large document corpora.

It composes:

• review-pdf for extractor quality/regression control on PDFs
• memory for graph ingestion
• taxonomy for federated bridge tags

Quick Start

cd ${HOME}/workspace/experiments/pi-mono/.pi/skills/learn-datalake

# One-shot run over a directory
./run.sh once /mnt/storage12tb/extractor_corpus --target-score 0.95