equinor/neqsim/paperlab-book-knowledge-graph

Skill: PaperLab Book Knowledge Graph

Purpose

Build a review graph for a PaperLab book so editors can see how chapters, figures, notebooks, standards, source folders, and skills connect. Use this when a book needs gap analysis, traceability review, duplicate-concept review, or a visual map of the editorial workflow.

Inputs

Use the book artifacts when they exist:

• book.yaml
• coverage_matrix.md
• source_manifest.json
• figure_dossier.json
• standards_map.json
• skill_stack_plan.json
• chapter notebooks and figures

Workflow

1. Run the prerequisite audits:

book-source-inventory, book-figure-dossier, book-standards-map, and book-skill-stack-plan.

1. Run python paperflow.py book-knowledge-graph <book_dir>.
2. Open book_knowledge_graph.html and review isolated nodes, missing skill

links, figures without discussion, chapters without notebooks, and standards that appear in text but have no clause-level evidence.

1. Use the graph during adversarial review to decide which chapters need

synthesis, source retrieval, or de-duplication.

Output

• book_knowledge_graph.json with nodes[] and edges[].
• book_knowledge_graph.html as a self-contained review page.

Review Questions

• Does every figure-heavy chapter connect to technical-figure-understanding

and figure-discussion?

• Do equipment chapters connect to standards and safety skills?
• Do notebook-heavy chapters have traceability markers?
• Are case-study chapters connected to source documents and skills?

Safety Rules

• Treat the graph as an editorial map, not proof of standards compliance.
• Do not expose private source documents in a public graph; include only path

labels and high-level source areas unless the source is public.

• Regenerate after chapter order, figure, source, or standards-map changes.