Pre-release Licensing Notice

1. Status of this notice

This page describes the current CohortDAG pre-release candidate. It is not a definitive End User License Agreement and is not a substitute for responsible legal review. The repository currently contains the BSD 3-Clause License, Copyright (c) 2025, Ono Lab. The installer reproduces that text verbatim as FIRST_PARTY_LICENSE.txt.

The ownership and licensing scope between that existing grant, the package's pre-release metadata, and proposed terms from Nexitia LLC remain under review. This notice does not alter or withdraw any existing licence grant, and it does not restrict rights supplied by third-party licences. Do not rely on this page as approval to redistribute a release candidate.

2. Pre-release status

CohortDAG is pre-release software. Features may change or be withdrawn, behaviour may change between builds, and there is no service level agreement and no uptime or data-preservation commitment. Do not rely on it as the system of record for anything you cannot afford to lose.

3. Not a medical device; not clinical advice

CohortDAG is a research and analysis tool. It is not a medical device and is not intended for the diagnosis, treatment, cure, mitigation, or prevention of any disease or condition. Its outputs — including any statistics, tables, curves, models, or figures — are not medical, clinical, diagnostic, or professional advice. You are responsible for independently validating every result and for any decision made on the basis of it.

4. The app runs locally

CohortDAG runs on your own machine. In normal operation it does not transmit your data, datasets, or analysis results to nexitia.ai or to the Licensor. Do not send protected health information (PHI), patient records, or other regulated or identifiable data to the hosted AROS Cloud Federation service; that prohibition is set out in the Terms of Service and mirrored here.

5. Managed upstream runtimes

The assisted macOS candidate reuses compatible native Java 17 and Python 3.12 installations or, after your explicit consent, downloads pinned native builds directly from the upstream projects into a per-user application-support directory. It verifies the expected size and SHA-256 before safe extraction and installs the exact Python analysis environment into an isolated virtual environment. The default path does not require Homebrew, administrator access, or shell-profile changes.

Those runtimes and packages — including Apache Spark (delivered by pyspark), pandas, scipy, statsmodels, lifelines, duckdb, sqlglot, pyarrow, pyreadstat, and openpyxl — remain third-party software licensed by their respective projects. They are obtained directly on your machine and are not embedded in the assisted CohortDAG installer.

Open-source components that the Licensor does redistribute inside the CohortDAG installer are listed, with their licences and required notices, in the Third-Party Notices.

6. Final terms are a release gate

Before public release, the responsible reviewer must approve the licensor and authority, first-party scope, treatment of existing grants, permitted use, privacy, support and update policy, warranty and liability language, termination, governing law, and acceptance/versioning. The approved terms must match the exact artifact and website notices. Until then, the candidate must not be promoted.

7. Contact

Licensing questions: nexitia@snow.ocn.ne.jp — 合同会社 Nexitia（ネクシア） / Nexitia LLC.