Shay b9a6f2ddb5 feat(packs): ADR-0100/0101/0102 — three sibling domain ratifications

Ratifies the remaining three sibling domains as reasoning-capable
under ADR-0091's Domain Pack Contract v1, using the template
ADR-0097 established for mathematics_logic. The capability ledger
now has four reasoning-capable rows backed by validated contracts.

ADR-0100 physics (en_physics_v1):
  domain_id: physics
  claimed_operators: causal, modal
  teaching_chains: [physics_chains_v1]
  eval_lanes: foundational_physics_ood, inference_closure,
    fabrication_control
  9/9 predicates pass

ADR-0101 systems_software (en_systems_software_v1):
  domain_id: systems_software
  claimed_operators: transitive, causal
  teaching_chains: [systems_software_chains_v1]
  eval_lanes: symbolic_logic, inference_closure, fabrication_control
  9/9 predicates pass

ADR-0102 hebrew_greek_textual_reasoning (FIRST MULTI-PACK ratification):
  domain_id: hebrew_greek_textual_reasoning
  claimed_operators: causal, contradiction
  teaching_chains: [hebrew_greek_textual_reasoning_chains_v1]
  eval_lanes: inference_closure, fabrication_control
    (universal lanes only — language-specific fluency lanes lack
    holdout splits; a separate ADR adds those when holdouts ship)
  packs: grc_logos_micro_v1, grc_logos_cognition_v1,
    he_logos_micro_v1, he_core_cognition_v1
  all four pack contracts identical (uniformity invariant pinned);
  all four 9/9 predicates pass
  pre-existing gap: hebrew/greek manifests lacked a provenance field
  entirely; ratification fills that uniformly across the four packs

44 new ratification tests in test_adr_0100_0102_sibling_ratifications.py:
- 6 parametrized 9-predicate validation tests (one per pack)
- 21 per-domain ledger status assertions (status, reasoning_capable,
  expert_demo gated, no_open_gaps, provenance points at correct ADR,
  operator_chain_coverage, intent_shapes minimum) — 7 cases × 3 domains
- 15 per-domain contract field shape assertions (teaching_chains,
  eval_lanes, splits coverage, axioms/rules null, primary reviewer) —
  5 cases × 3 domains
- 2 ADR-0102 multi-pack uniformity invariants (all four packs carry
  the contract; contracts identical across packs)

Capability ledger after ratification:
  systems_software           : reasoning-capable
  mathematics_logic          : reasoning-capable
  physics                    : reasoning-capable
  hebrew_greek_textual_reasoning : reasoning-capable
  philosophy_theology        : reasoning-capable (no contract; pre-existing)

Lane SHA pin update:
- public_demo pin refreshed (21751aaf.. → 71090323..) — the
  ratification adds new manifest fields (provenance,
  domain_contract_*) that surface in pack-related demo paths;
  intentional ADR-tracked change per the verifier doctrine

Smoke 67/67, packs 6/6, sibling ratifications 44/44, cognition eval
byte-identical 100/100/100/100; all 6 lanes match pinned SHAs:
  reviewer_registry            681a2aab..
  miner_loop_closure           9f071733..
  domain_contract_validation   f9c06cde..
  fabrication_control_summary  01e1b6b7..
  demo_composition             27d83824..
  public_demo                  71090323..

2026-05-21 20:25:48 -07:00

3.4 KiB

Raw Blame History

ADR-0101 — Systems-Software Reasoning-Capable Ratification

Status: Proposed Date: 2026-05-21 Author: CORE agents + reviewers Depends on: ADR-0091, ADR-0092, ADR-0093, ADR-0096, ADR-0097

Context

The systems_software substrate is structurally complete:

language_packs/data/en_systems_software_v1/ ships with manifest, lexicon, glosses.
teaching/domain_chains/systems_software_chains_v1.jsonl exists.
evals/symbolic_logic/ exists with dev/public/holdouts (the closest in-tree lane for systems/software reasoning at v1).
docs/gaps.md marks every systems_software gap closed: gap:systems_software_pack_absent, gap:systems_software_transitive_chains_below_threshold, gap:systems_software_causal_chains_below_threshold, gap:systems_software_intent_shapes_below_threshold — all [x].
Chain coverage: causal=8, transitive=8 (≥8 per claimed operator family ✓); five intent shapes populated (≥3 ✓).

What's missing is the ADR-0091 formal ratification step, matching the template established by ADR-0097 for mathematics_logic.

Decision

Ratify en_systems_software_v1 as reasoning-capable under ADR-0091 by emitting the contract fields into its manifest and passing core capability domain-contract validate en_systems_software_v1.

Manifest additions

{
  "domain_contract_version": 1,
  "domain_id": "systems_software",
  "axioms": null,
  "rules": null,
  "teaching_chains": ["systems_software_chains_v1"],
  "eval_lanes": [
    {"lane": "symbolic_logic", "version": "v1",
     "splits": ["dev", "public", "holdout"]},
    {"lane": "inference_closure", "version": "v1",
     "splits": ["dev", "public", "holdout"]},
    {"lane": "fabrication_control", "version": "v1",
     "splits": ["dev", "public", "holdout"]}
  ],
  "reviewers": ["shay-j"],
  "known_gaps": [],
  "provenance": "adr-0101:reviewed:2026-05-21"
}

Claimed operator families (DOMAIN_OPERATOR_CLAIMS): transitive, causal. axioms and rules stay null at v1 — systems-software reasoning is demonstrated through reviewed causal/transitive chain composition over invariants, boundaries, and dependencies.

Eval lane choice — `symbolic_logic`

Systems-software lacks a dedicated v1 evaluation lane. symbolic_logic is the closest in-tree fit: it pins reasoning over typed relations (implication, contradiction, dependency) that the systems_software chain corpus exercises. A future ADR may introduce a systems_software_design_ood lane and supersede this declaration.

Invariant

systems_software_reasoning_capable_ledger_row — core capability ledger emits a row for domain_id: systems_software with status: reasoning-capable, provenance pointing at this ADR, and expert_demo: False until a future ADR attaches the required reports.

Lane

No new lane. Existing lanes carry the evidence:

symbolic_logic/ (reasoning over typed relations)
inference_closure/ (composition)
fabrication_control/ (negative control, ADR-0096)
evals/domain_contract_validation/ (ADR-0093 confirms predicates fire)

PR Checklist

Capability added: third ratified reasoning-capable domain.
Invariant proved: systems_software_reasoning_capable_ledger_row.
Hidden normalization / stochastic fallback / approximate recall / unreviewed mutation: none.
Trust boundary: manifest edit through reviewed flow; checksums unchanged.

3.4 KiB Raw Blame History