V2 implementation substrate for REE claim experimentation. Upgraded from synthetic scaffolding (2026-02-26 archive) to a real ree_core implementation as of 2026-03-06.

• HippocampalModule (new in V2): terrain-navigated trajectory proposal. Resolves SD-001 — CEM-based trajectory search was misplaced in E2; it is now here, using E2 only as a forward rollout model.
• E2: pure fast transition model f(z_t, a_t) → z_{t+1}. No longer performs candidate generation.
• CausalGridWorld: real environment replacing synthetic data generation.
• SD-002 resolved (2026-03-06): E1 prior wired into HippocampalModule's terrain search (E1→HippocampalModule mutual constitution).
• SD-003: forward_counterfactual() exposed via E2 for self-attribution. Substrate ready; experiments pending.
• SD-004 (held for V3): Action objects as hippocampal map backbone — E2 produces compressed action-object representations that HippocampalModule maps over instead of raw state space, enabling much longer planning horizons. See docs/architecture/design_decisions.md.

V2 series complete: 15 experiments run (EXQ-014–EXQ-028) against real ree_core — 6 PASS, 7 FAIL (EXQ-027/028 were SD-003 scoping experiments). All three V2 hard-stop criteria met; V3 transition formally triggered. Governance sync complete: results indexed in REE_assembly/evidence/experiments/ via sync_v2_results.py + build_experiment_indexes.py. Full results in evidence/experiments/. Historical queue in experiment_queue.json.

Run full ree-v2 <-> REE_assembly flow (qualification, handoff sync/ingestion, dispatch emission, dispatch pullback):

scripts/cross_repo_roundtrip.sh

Preview commands without executing:

scripts/cross_repo_roundtrip.sh --dry-run

Apache License 2.0 (see LICENSE).

• Cite this repository using CITATION.cff.
• For canonical architectural attribution, cite Daniel Golden's REE specification in https://github.com/Latent-Fields/REE_assembly/ (also captured as the preferred citation in CITATION.cff).