anatomy-graphdb 1.0.0

Canonical anatomy domain dataset and Neo4j graph builder for Stronger.

Stronger Anatomy (anatomy-graphdb)

Canonical anatomy domain package for Stronger.

This repository provides:

• structured YAML anatomy data in config/
• typed Python loaders/models in anatomy_graphdb
• canonical catalog exports for downstream domains (anatomy_graphdb.exports)
• a CLI (anatomy-graphdb) for validation and Neo4j export/ingestion
• packaged cleaned overlay SVG assets in anatomy_graphdb/assets/

Install

Minimal package (no Neo4j driver)

pip install anatomy-graphdb

With Neo4j ingestion support

pip install "anatomy-graphdb[neo4j]"

Local development

poetry install

Standalone Usage

Load anatomy regions in Python

from anatomy_graphdb.databases.anatomy.loader import AnatomyLoader
from anatomy_graphdb.domain import build_anatomy_model

loader = AnatomyLoader()
region = loader.load_region("chest")
model = build_anatomy_model(region)

print(model.region)
print(len(model.muscles))

Access packaged overlay assets

from anatomy_graphdb.assets import load_overlay_manifest, asset_absolute_path

version, assets = load_overlay_manifest()
front_outline = assets["__base__"]["front"]
path = asset_absolute_path(front_outline)
print(version, path)

CLI

List available regions:

poetry run anatomy-graphdb --list-regions

Validate a region:

poetry run anatomy-graphdb --region chest --validate-only

Export CSV artifacts:

poetry run anatomy-graphdb --region all --output data/neo4j --validate

Export canonical anatomy catalog JSON:

poetry run anatomy-graphdb --export-catalog --region all --catalog-output data/catalog/anatomy_catalog.json

Direct Bolt ingestion (requires neo4j extra):

poetry run anatomy-graphdb --region all --mode bolt --validate

Integration Pattern (Your Multi-Repo Setup)

Recommended layering:

1. stronger-domain-postgres: relational exercise/workout domain.
2. anatomy-graphdb (this repo): anatomy graph domain + canonical overlay assets.
3. stronger-api: composition layer that imports both domain packages.
4. stronger-frontend: consumes API JSON + SVG asset URLs.

In your current setup, stronger-api imports this package and serves:

• anatomy graph data from YAML-derived models
• overlay manifest and SVG files from anatomy_graphdb/assets
• catalog/version metadata for drift debugging

stronger-domain-postgres can consume --export-catalog output (or import exports directly) to seed canonical muscle_groups and muscles without duplicating anatomy configs.

The frontend should consume those API routes instead of maintaining duplicated anatomy SVG sources.

Asset Layout

Canonical packaged assets:

• Manifest: anatomy_graphdb/assets/overlay_manifest.json
• SVG files: anatomy_graphdb/assets/anatomy/*.svg

Legacy repo-only assets (not packaged into wheels):

• svgs/svg_front_muscles
• svgs/svg_rear_muscles
• svgs/muscles
• svgs/manifest.json

Those svgs/ folders are retained for historical/reference tooling only.

Neo4j Configuration

When using bolt mode, configure environment variables:

NEO4J_URI=bolt://localhost:7687
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=password
NEO4J_DATABASE=neo4j

For AuraDB:

NEO4J_URI=neo4j+s://<instance-id>.databases.neo4j.io
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=<your-password>
NEO4J_DATABASE=neo4j

Then run:

poetry run anatomy-graphdb --region all --mode bolt --validate

Docker Graph Build

You can run Neo4j in Docker and build/populate the graph in one command:

make neo4j-refresh-docker REGION=all

Useful helpers:

make neo4j-up
make neo4j-logs
make neo4j-down

Override defaults when needed:

make neo4j-refresh-docker REGION=chest NEO4J_AUTH=neo4j/password HTTP_PORT=7475 BOLT_PORT=7688

Tests

poetry run pytest

Public API and Versioning

• Stable API surface: docs/API_STABILITY.md
• Changelog: CHANGELOG.md
• Release process: docs/RELEASING.md
• Asset attribution policy: docs/ASSET_ATTRIBUTION.md

Package Metadata

• Package name: anatomy-graphdb
• Python: 3.12+
• License: MIT