Current version
CloudCaive Assessment Rubric v1.0 is the current rubric. It defines what an assessed session is, what is sampled, what evidence is reviewed and how, the three possible outcomes for each sampled objective, and the limits of what a report may claim.
Reports identify the rubric version they were graded against, for example: “Assessed using CloudCaive Rubric v1.0.”
How versioning works
- Immutable versions. Once a rubric version is effective, no semantic change is made at its address: any substantive change produces a new version at a new address. Purely presentational or accessibility repairs that leave meaning unchanged may be applied, and each is recorded in the change history.
- Effective dates are publication dates. A version becomes effective on the date it first becomes publicly accessible in production; until then it is a draft and no report may cite it.
- Stable addresses. Superseded versions remain published at their original address so older reports stay verifiable.
- No mid-cycle changes. A participant’s baseline and reassessment are always graded against the same rubric version. A rubric is not changed part-way through anyone’s baseline-to-reassessment cycle.
- Every report cites its version. The pinned rubric version and scenario version are recorded on every report.
Version history
| Version | Effective date | Status | Change summary |
|---|---|---|---|
| v1.0 | 4 August 2026 | Current (effective) | Initial published rubric for the founding beta. |
What this page is not
The rubric is the public statement of how evidence is judged. The assessment incidents themselves, their benchmark answers and the form-comparison material for unseen reassessment forms are deliberately held back — publishing them would let a session be rehearsed rather than sampled, and would break the blind reassessment. The rubric explains how judgement is made; it does not reveal the specific items.