cdo-review

Run the AI training data audit script with a JSON file of your data sources to classify each as GO, MITIGATE, or NO-GO based on consent provenance. The review checks origin, consent flow, data class, and intended use for every source.

The choice depends on internal consumer count: under 5 consumers favors a warehouse, 5-25 favors a lakehouse, and 25+ with federated culture favors mesh. The data product strategy picker script formalizes this decision from a profile JSON input.

Run it before approving ML training on customer data, signing multi-year data infrastructure contracts, productizing customer data, making major data hires, or entering M&A diligence. It also applies whenever data monetization is proposed.

Yes, it checks anonymization documentation, MSA carve-out coverage, and training-data provenance logs. The data asset valuator script estimates strategic value, moat strength, and ARR multiplier ranges for the data corpus.

It is a decision framework, not a legal opinion, so productization or licensing paths still require routing to general counsel review. It also depends on accurate JSON inputs describing data sources and organizational profile.