score

Version 0.1 · Open standard

An open standard for AI knowledge management.

Your AI knowledge, independent of any vendor.


Why Score

Know

What your AI has access to.

Govern

What it is permitted to say and do.

Audit

What it actually said and when.


The problem

Every AI deployment has a knowledge problem. What the model can access, what it is permitted to say, and what it actually said are three distinct questions. Most organisations answer none of them clearly. Score is a common format for answering all three, so your deployment can be governed rather than guessed at.


Two paths

For senior leaders and enterprise teams

If you are thinking through what Score means for your organisation's AI strategy, MultipleWorks is available for advisory conversations.

Get in touch →

For developers

The full specification, documentation, and examples are in the GitHub repository.

View the spec →