Install¶

Three supported installation paths, in order from "least effort" to "most customisable".

1. Docker (fastest path to running)¶

The published image at ghcr.io/synaptixs/ontomesh:3.9.0 is multi-arch (linux/amd64 + linux/arm64), runs as a non-root user, and bundles every default extra needed to open the wizard.

# Ephemeral (resets when the container stops):
docker run --rm -p 5051:5051 ghcr.io/synaptixs/ontomesh:latest

# Persistent (saved ontologies survive restarts):
docker volume create ontomesh-data
docker run -d --name ontomesh \
  -p 5051:5051 \
  -v ontomesh-data:/data \
  -e ONTOMESH_DATA_DIR=/data \
  ghcr.io/synaptixs/ontomesh:latest

Then open http://localhost:5051.

Verify what you pulled¶

The image is cosign-signed by GitHub Actions OIDC — anyone can verify, offline:

cosign verify ghcr.io/synaptixs/ontomesh:3.9.0 \
  --certificate-identity-regexp='^https://github.com/synaptixs/ontomesh' \
  --certificate-oidc-issuer=https://token.actions.githubusercontent.com

The image also ships an SPDX SBOM and SLSA provenance attestation as OCI artefacts:

docker buildx imagetools inspect ghcr.io/synaptixs/ontomesh:3.9.0 \
  --format '{{ json .SBOM }}'

2. From PyPI¶

Published to PyPI as ontoforge from 3.8.0.

pip install ontoforge                 # core only — small footprint
pip install 'ontoforge[wizard]'       # adds Flask + the wizard UI
pip install 'ontoforge[wizard,postgres,runtime]'   # production-ish
ontoforge-wizard                       # launches the wizard

3. From source (for contributors)¶

git clone https://github.com/synaptixs/ontomesh.git
cd ontomesh
python -m venv .venv && source .venv/bin/activate
pip install -e ".[wizard,test]"
pytest tests/ -q              # 297+ tests should pass
ontoforge-wizard               # starts the wizard at :5051

Optional extras¶

The package ships with a set of optional extras you can compose:

Extra	Adds	When you need it
`wizard`	Flask, flask-cors, gunicorn, jsonschema, prometheus-client, python-json-logger	Running the wizard UI
`postgres`	psycopg2-binary + psycopg3	Postgres backend for the saved-ontologies store
`mysql`, `mssql`, `oracle`, `db2`	Respective drivers	Schema introspection against those engines
`runtime`	anthropic, openai, google-cloud-aiplatform, requests, oci	LLM adapters used at runtime
`mining`	drain3, hmmlearn, statsmodels, scikit-learn	Log mining phases (L4–L13)
`drift`	drain3 (plus a follow-on `pip install git+https://github.com/synaptixs/infodrift.git` for `drift-monitor` itself)	Production drift monitoring
`discover`	spaCy	Log-entity discovery
`neptune`	boto3	AWS Neptune graph backend
`redis`	redis-py	Multi-worker SSE bus
`test`	pytest	Running the test suite
`all`	every runtime/drift/wizard/db extra	Convenience aggregate

For the full list see pyproject.toml.