This document defines the current operational phase of the matrix project.
It does not redefine matrix. It describes the present testing posture.
To test whether the admissibility and STOP framework produces stable and consistent diagnostic outcomes when applied to minimal real content.
This phase does not aim to expand knowledge coverage. It aims to stress-test the admissibility engine.
- Exactly 2 domains selected
- 4 diagnostic runs total:
- 2 STOP cases
- 2 bounded ALLOW cases
- No additional domains during this phase
- No modification of core admissibility rules unless forced by structural failure
Default assumption for legacy-derived claims: inadmissible.
ALLOW must be explicitly justified and bounded.
This phase is considered complete when:
- 2 domains are documented with clear rationale
- 4 runs are fully documented and internally consistent
- At least one recurring structural friction or failure mode is identified
- No core rule modifications are required by content
- A written Month-1 Review evaluates scalability and structural stress
- No domain expansion beyond the 2 selected
- No claim of truth, correctness, or authority
- No conversion of legacy material into canonical matrix content
- No public positioning as a knowledge archive
Once evaluation criteria are met, a decision will be made to:
- Continue controlled solo expansion (estimated 10–15 domains feasible in 2026), or
- Prepare structured contributor onboarding, or
- Further narrow scope intentionally
Until that decision, expansion is suspended.
Legacy material remains archived and non-operative.
This phase tests the admissibility engine against selected cases. It does not rehabilitate or migrate historical material.
Status: Active
Phase Start: February 2026