Skip to content

[AIAA] Assessment-quality rubric (pre-pilot gate) #381

Description

@nate-double-u

Pre-pilot gate (spec sections 10, 16, 17), not a numbered build step:
the assessment-quality rubric must exist before the pilot's first
deliverable is scored. A pilot scored without it leaves G-2 asserted,
not verified.

Done when:

  • The meta-rubric that scores an assessment's quality is defined,
    distinct from criteria.md, which scores a project's docs.
  • It is validated by scoring the Flatcar, Knative, and Helm
    baseline assessments to set a reference band (sections 2, 10).

This work also sets any post-pilot sampling rule for verification,
replacing the pilot's check-every-finding floor (section 10).

Metadata

Metadata

Assignees

No one assigned

    Labels

    CI/infraCI & infrastructure

    Type

    No type

    Projects

    Status
    Todo

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions