Shay
|
6207b5fd0e
|
feat(register): R1–R4 register pack subsystem — deterministic surface variation
Introduces the presentation axis as a fourth pack class (sibling to identity /
safety / ethics), orthogonal to the truth path. Same input + same packs +
same register ⇒ bit-for-bit reproducible surface; varying any of the three ⇒
genuinely different output. No stochastic sampling.
ADR-0068 (R1): RegisterPack frozen dataclass, loader, ratify script, seam test.
- default_neutral_v1 ratified as null register.
ADR-0069 (R2): realizer register parameter threaded through 9 composer entry
points; RuntimeConfig.register_pack_id; three byte-identity invariants
(A: None ≡ pre-R2 unregistered; B: None ≡ default_neutral_v1; C: trace_hash
invariant under register). Amended to default-with-lint after 167-call-site
scout: composers default to UNREGISTERED, AST lint enforces explicit
register= at runtime call sites.
ADR-0070 (R3): terse_v1 register, first non-neutral pack. realizer_overrides
schema with known-keys allow-list (disclosure_domain_count ∈ {1,2,3}).
build_pack_surface_candidate reads override with fail-soft clamp. New
invariant register_invariant_grounding asserts grounding_source +
trace_hash byte-identical across {None, neutral, terse}.
ADR-0071 (R4): seeded surface variation via convivial_v1.
chat/register_variation.py applies SHA-256-seeded marker selection from
bounded discourse-marker buckets. ChatResponse.pre_decoration_surface routes
truth-path surface to core/cognition/pipeline.py so trace_hash stays
invariant under register (the load-bearing architectural fix — initially
invariant C failed under convivial because decoration was leaking into
trace_hash via response.surface). Empty-string marker entries now
legitimate ("no marker this turn" is a valid seed pick). realizer_overrides
schema widened with per_intent nested block (validated against IntentTag
whitelist; wired but not exercised by convivial). Two new invariants:
seeded_variation_replay_equivalence (fresh runtimes → byte-identical) and
seeded_variation_turn_distinct (same prompt across turns → ≥2 distinct
surfaces).
ADR-0072 (R5, draft): telemetry + operator surface — TurnEvent gains
register_id and register_variant_id, core chat --register flag, core demo
register-tour. Status: Proposed; not yet implemented.
Three ratified register packs ship: default_neutral_v1 (null), terse_v1
(disclosure_domain_count=1), convivial_v1 (3 openings × 3 closings).
Verification:
- 84 register tests pass + 1 documented skip
- Curated lanes green: smoke 67, cognition 120+1s, teaching 17, packs 6,
runtime 19, algebra 132
- Cognition eval byte-identical to pre-register baseline:
public 100/100/91.7/100, holdout 100/100/83.3/100
- Full lane: 2608 passed, 4 skipped, 1 failed (pre-existing
test_cli_demo.py "Combined Demo" → "Run Every Demo" rename, unrelated)
Truth-path isolation: chat/register_variation.py is realizer-side; the seam
test (tests/test_register_pack_seam.py) refuses imports of packs.register
from intent classification, propagation, vault recall, trace hashing, and
algebra.
|
2026-05-19 16:52:36 -07:00 |
|
Shay
|
d5a6e81b33
|
feat(adr-0067): cross-pack teaching chains — Plan Phase 4 closed
ADR-0064 bound each teaching corpus 1:1 to a single ratified pack;
chains whose subject + object resolved to different packs were
dropped at load time. Phases 1–3 ratified the per-pack DAGs needed
to lift that constraint safely.
ADR-0067 introduces a deliberately narrow cross-pack chain shape.
Each entry carries explicit subject_pack_id and object_pack_id
fields, and the loader verifies per-chain residency. Same-pack
entries are rejected as corpus-misfilings (anti-leakage). The
cross-pack composer is the fall-through after the in-pack composer,
so the cognition lane stays byte-identical.
Files:
- chat/cross_pack_grounding.py — CrossPackChain + loader +
single-chain composer + multi-chain enumerators
- teaching/cross_pack_chains/cross_pack_chains_v1.jsonl — 5 seed
chains (family×identity, parent×understanding, family×memory,
identity×family, understanding×parent)
- chat/runtime.py — fall-through wiring in CAUSE/VERIFICATION
- chat/narrative_surface.py, chat/example_surface.py — merge
cross-pack chains, per-chain pack-residency helpers
- tests/test_cross_pack_chains.py — 31 tests covering loader,
surface, multi-chain access, runtime integration, in-pack
precedence
- tests/test_narrative_example_intents.py — corpus-tag assertions
widened to allow cross-pack aggregation
Verification:
- 31 new tests pass
- Curated lanes: smoke 67 / cognition 121 / teaching 17 / packs 6 /
runtime 19 — all green
- Cognition eval byte-identical (public 100/100/91.7/100, holdout
100/100/83.3/100)
- Full lane: 2098 passed, 2 skipped, 0 failed in 2:30
|
2026-05-18 17:22:43 -07:00 |
|