55a3f79f18
Two prior arcs closed on "the binding constraint is ratified curriculum volume, not machinery." Verified against code 2026-07-25: the machinery half is wrong in a load-bearing way, and the correction is the reason this arc exists. evals/curriculum_serve/practice/ does not exist. seal_ledger exists only for deduction (evals/deduction_serve/practice/runner.py:96), yet chat/curriculum_serve_license.py's own docstring names evals.curriculum_serve.practice.runner.seal_ledger as "the only writer" of the artifact it reads -- that writer was never built, and chat/data/curriculum_serve_ledger.json does not exist. And the ceremony stops short: RatificationReceipt.pending_stages = (arena_queue_entry, ledger_reseal). So the pipeline is author -> ratify -> ??? -> ??? -> license -> serve. Authoring curriculum today cannot earn a license at ANY volume. Content is necessary and currently insufficient. The build is still modest, because oracle.py is already an independent decision procedure usable as the gold tether and ADR-0263 already extracted the sealers for exactly a third instance. Two measured facts shape the content work. `refuted` has NEVER been produced -- the physics lane gold is entailed 14 / unknown 12 / declined 6, and no corpus row can express it. And bands key on (domain, operator_family), so negative content must live INSIDE operator_family "modal" with a row-level polarity marker; a separate modal_negative family would create a new band at n=0 instead of adding refuted volume to the target band -- the easiest available way to waste the whole authoring effort. DIVISION-OF-WORK.md records the tiering criterion and why it is not difficulty. The evidence does not support a prestige split: in the same 48h window Sonnet 5 found the DEFAULT_SINK bug that wrote outside the repository and did the six-worktree public_demo archaeology, while Opus 5 mis-diagnosed the CGA hot path by multiplying a microbenchmark by a call count. Model tier was not the variable; verification discipline was. The split is therefore: would an error here be caught by a gate, or is it silent? Two units are silent-failure BECAUSE THEY ARE THE GATE -- the negative-curriculum epistemology (wrong => corpus encodes falsehoods and wrong=0 still passes, since lane gold and the oracle both read the same rows) and the volume-honesty invariant (wrong => conservative_floor cannot tell 657 independent facts from 16 asked 41 times, so the ledger clears theta=0.99 having earned nothing). Those get Opus. Everything loud goes to the cheapest competent tier. Also records the ordering correction: the epistemology ADR must precede the generator, or the generator gets built twice. [Verification]: docs-only; no code paths touched. Every referenced path confirmed to exist and all five line citations confirmed accurate at HEAD (gold.py:1163, curriculum_serve/runner.py:71, discovery_yield.py:67-70, entail.py:125, curriculum_surface.py:184). Pre-push gate runs on push. |
||
|---|---|---|
| .. | ||
| adr | ||
| agents/grok | ||
| analysis | ||
| architecture | ||
| audit | ||
| benchmarks | ||
| briefs | ||
| curriculum | ||
| decisions | ||
| evals | ||
| examples | ||
| handoff | ||
| handoffs | ||
| implementation | ||
| issues | ||
| lab | ||
| outreach | ||
| paradigm-archive | ||
| plans | ||
| research | ||
| sessions | ||
| specs | ||
| workbench | ||
| zig | ||
| 3lang-depth-pr-plan.md | ||
| admissibility-exemplars.md | ||
| ci-optimization.md | ||
| core-rd-base-prompts.md | ||
| ethics_packs.md | ||
| EVAL_AUDIT_2026-05-20.md | ||
| eval_methodology.md | ||
| frontier_baselines.md | ||
| gaps.md | ||
| handoff_template.md | ||
| hitl-backpressure.md | ||
| holdout_recipients.txt | ||
| identity_packs.md | ||
| master-plan-post-substrate-audit.md | ||
| memo.html | ||
| model_dependency_size_tally.md | ||
| pack_inventory_2026-05-21.md | ||
| position_paper.md | ||
| PROGRESS.md | ||
| README.md | ||
| recognizer-registry.md | ||
| refusal-taxonomy.md | ||
| reviewers.yaml | ||
| RUST.md | ||
| safety_packs.md | ||
| sponsors.md | ||
| teaching_order.md | ||
| test-debt-quarantine.md | ||
| testing-lanes.md | ||
| Whitepaper.md | ||
| Yellowpaper.md | ||
CORE Documentation Index
This is the central index for all documentation in the CORE project.
Canonical Root Documents
- Whitepaper.md - The CORE architectural and philosophical whitepaper.
- Yellowpaper.md - Technical specifications and mathematical formulation of the CORE engine.
- PROGRESS.md - High-level project progress tracking.
- specs/runtime_contracts.md - Critical invariants and bounds for the runtime execution flow.
Directories
Architecture & Design
- adr/ - Architecture Decision Records (ADRs). The canonical history of all ratified engineering choices.
- architecture/ - High-level architectural documents (e.g., pipelines, schemas).
- analysis/ - Deep dives and master plans for structural changes.
- specs/ - Detailed technical specifications and invariants.
Planning & Progress
- plans/ - Capability roadmaps and implementation plans.
- briefs/ - Project briefs and scoping documents for upcoming work.
- issues/ - Detailed issue analyses and technical reproductions (not standard trackers).
- audit/ - Audit reports and claims ledgers.
Operations & Usage
- examples/ - Usage examples and reference integrations.
- workbench/ - Documentation for the CORE workbench UI and related tooling.
- agents/ - Agent-specific operational guides.
Note regarding
agents/grok/: Operational guide for using Grok 4.3 + Grok Build with CORE. This is not architecture documentation.
Learning & Evaluation
- curriculum/ - Documentation on the teaching/learning order and knowledge progression.
- evals/ - Evaluation methodology and performance criteria documentation.
- benchmarks/ - Benchmark evidence and performance tracking logs.
Historical & Experimental
- sessions/ - Chronological session logs documenting the "decision trail" and intellectual history of major choices.
- handoffs/ - Legacy brief, audit, and investigation notes (historical; the formal HANDOFF mechanism is retired — see AGENTS.md for the current lightweight
session-break-summary-<DATETIME>.mdconvention). - research/ - Raw research notes and preliminary findings.
- lab/ - Experimental content (Warning: Not ratified; must not be referenced as authoritative in production PRs).