ns-pond-ingestion-workflow

Finds neuroimaging papers, downloads them from whichever source will serve them, pulls the coordinate tables out, turns those into analyses, and writes the result to Neurostore and to ns-pond.

The system as it stands

Design	identity, fingerprints, stages, the command line, choosing a deployment
Data safety	how migration avoids losing the corpus
Migration	command-by-command upgrade guide

These describe the pipeline as it is now. Each change edits them.

Changes

Records of changes worth explaining — what they replaced, what they measured, and the reasoning. Frozen once merged.

• #11 Catalog refactor — replaced the six stage caches with one article catalog (before · benchmarks · profiling)

Running it

See the README for installation and the command line.