Shay bdb294eac3 feat(workbench): land B3.5-b/c/d/e — calibration evidence subject, B4a leeway gate, docs; runner-reproducible practice artifact

Completes the Wave M B3.5 consolidation slice (b–e), built on #728.

B3.5-b — calibration as a first-class evidence subject (`calibration_class`,
address `calibration:<class_name>`): RightInspector projection + Evidence
Chain Rail semantics (serving-discipline evidence, not runtime truth).

B3.5-c / B4a — nullable `LeewayEvidence` read model threaded through turn,
replay, cognition-proposal, and math-proposal surfaces, with a shared
absence-honest card. B4 is gated correctly: the tuple exists in typed data but
no producer populates it, so the card renders absence (verified: no non-null
producer in workbench/core/chat).

B3.5-d/e — UI-UX-GUIDE.md, b4-leeway-feasibility-gate.md, phase-a-residue-ledger.md.

Practice artifact — earn-it-for-real (runner-reproducible). The committed
`report.json` (additive earns PROPOSE @0.861, 95/5/50) is now emitted by a
deterministic runner rather than copied from the queue. `propose_runner`
gains `regenerate_practice_artifacts()`, which runs ONE sealed `resolve_pooled`
practice pass and writes BOTH report.json (the per-class ledger the calibration
reader consumes) and ratification_queue.json — two projections of one ledger,
coherent by construction and byte-reproducible. `runner.main()` delegates to
it (lazy import, no cycle), so both entry points produce the identical pair.
This closes the gap where a hand-copied report.json agreed with the queue but
no runner produced it. `resolve_pooled` is the aggressive sealed PROPOSE-regime
scorer (proposal-only/HITL, unsafe for serving, legitimate for
attempt-and-eliminate); wrong=5 is the sealed-practice learning signal, NOT the
serving wrong=0. No serving/derivation/reliability_gate source touched; the
practice lane is not in the serving-frozen SHA gate.

Validated:
- python -m pytest tests/test_workbench_{calibration,journal,replay,schemas}.py -> 31 passed
- python -m pytest tests/ -k "workbench or propose or learning_arena or practice"
  -> 190 passed (3 failing tests in test_adr_0175_phase2_practice_lane.py are
  PRE-EXISTING reds on clean origin/main: stale 4/0/46 assertions on build_report,
  which this change does not touch)
- report.json + ratification_queue.json: deterministic (run1==run2) and
  reproduced byte-identically by both `python -m ...runner` and `...propose_runner`
- pnpm build green; 144 UI tests across calibration/leeway/evidence/replay/
  doctrine-gates/routes-docs-drift all pass

2026-06-13 07:36:44 -07:00

1.6 KiB

Raw Blame History

Wave M B3.5-c — B4 Leeway Feasibility Gate

Date: 2026-06-13 Status: B4a read model created; full B4 annotations not yet admitted

Finding

B4 needs a tuple that explains why approximation or leeway was granted:

class_name
license: PROPOSE | SERVE | blocked | unknown
theta
claim_disclosure: approximate | verified | proposal_only | none
source_digest
calibration_evidence_ref

Before this slice, that tuple was not present in ChatTurnResult, TurnJournalEntry, ProposalDetail, or MathProposalDetail. Any UI card that named class/license/theta would have had to infer calibration state in the frontend.

B4a Read Model

workbench/schemas.py::LeewayEvidence is now the only lawful tuple shape for B4 explanations. It is nullable on:

ChatTurnResult
TurnJournalEntrySchema
TurnReplayComparison
ProposalDetail
MathProposalDetail

Frontend mirrors live in workbench-ui/src/types/api.ts. The shared LeewayEvidenceCard renders the tuple when present and renders explicit absence when it is null or missing.

Gate Result

B4 UI annotations may start only when a backend producer populates LeewayEvidence from engine-owned calibration evidence or a lawful backend join. Until then, Proposals and Replay can show "No leeway evidence recorded."; they must not explain leeway from frontend-only class/license inference.

Follow-Up

Next PR: B4 producer wiring. Candidate sources are the turn/proposal creation paths that already know whether a claim is approximate, plus the calibration reader's calibration:<class_name> evidence subject.

1.6 KiB Raw Blame History