Consolidation before application. Everything TD-WP-0002 demonstrated rests on
one use case, one application and one token-free runtime; three of the
project's own findings say that is too narrow a base for real work.
- T01 bounded live-model experiment: settles F-0005 (capability) and F-0007
(economics) in one run. Needs an operator decision on cost ceiling first.
- T02 two further use cases: the model was designed against Alice/Bob/Carol,
so of course it fits. What it has to grow to express something else is the
evidence.
- T03 build the dropped-identifier mutation side to ten; H-001 currently
turns on three.
- T04 settle every gated concept - Temperature and energy.py are used in a
decision or removed. Extending a gate is not a result.
- T05 decide how claims are expressed, answerable only after T02.
- T06 browser-engine driver if warranted (blocked on Playwright).
- T07 measure time to express a use case - first in the scorecard, never
measured, and reconstructable only during T02.
- T08 readiness review with explicit criteria for real-system application.
Applying test-driver to audit-core is explicitly out of scope and gated on
T08. Coordination note added: another session works in this repo, so staging
must be explicit and interface changes published.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1629012@bnt-lap001
Assistant-Session: 78d4fb13-8a1e-474b-87a3-9b9261c49a39
tegwick
faef069b17