Engineered Water Systems knowledge graph

This repository contains one production paper-to-knowledge-graph pipeline, its source papers, generated audit boundary, and local Neo4j runtime.

Start here

The complete dependency, local-model, configuration, and operating guide is Pipeline/README.md.

For a Windows-first, step-by-step operating runbook covering the website, direct pipeline runs, Neo4j startup, and graph viewing, use STANDARD_OPERATING_PROCEDURE.md.

Quick sequence after following that setup guide and starting the configured Neo4j instance:

python Pipeline/check_environment.py --mode full
python -m Pipeline.GENERAL.node_only_server

Open http://127.0.0.1:8767/.

Folder map

Folder	Purpose
Pipeline/	Canonical implementation, runtime assets, requirements, tests, and setup guide
SamplePapers/	Workbench, evaluation, and regression source PDFs
PipelineAudits/	Retained generated runs and review evidence
LocalNeo4j/	Local Neo4j distribution and paper-graph instance

Do not create new root-level output, temporary, or report folders. Pipeline output belongs under PipelineAudits/.