A verification-first research programme toward a unified field theory.
Maintainer: Jusang Lee (jtkor@outlook.com) · Bootstrapped: 2026-06-05
TECT is operated as a Unified Classical Field Theory (UCFT) / partial-TOE research
programme. No TOE-level claim is made at the programme level; the TOE statement is
managed as an explicit Master Theorem decomposed into sector theorems with a
tracked dependency DAG (see GOVERNANCE.md §1). Every result is registered as a
claim card with a precise statement, pinned scope, dependency list, evidence
grade, falsification gate, and maturity tier. Nothing in this repository should be
read as stronger than its registered tier.
The operating principle of this repository:
Do not make TECT look complete. Make TECT impossible to misunderstand.
| Path | Content | Publication tier |
|---|---|---|
CLAIMS.md |
Master claim ledger (auto-generated — start here) | P1 |
CATALOG.md |
Every artefact with versions, dates, claim links (auto-generated) | P1 |
RESULTS-LEDGER.md |
Standalone-publishable results harvested from the claims (curated) | P1 |
strategy/ |
Non-tier-bearing strategy / analysis / decision-rationale notes (route planning, impact studies) | P1 |
ROADMAP.md |
6-Stage roadmap v2 with exit conditions and current status | P1 |
GOVERNANCE.md |
Operating constitution (tiers, gates, registration rules) | P1 |
REVIEWING.md |
How to review or attack TECT in 30 minutes | P1 |
governance/ |
Detailed binding policies | P1 |
claims/ |
One folder per claim: card + status.json + proof notes (notes/) + run artefacts (runs/) + development lineage (LINEAGE.md, generated; lineage-narrative.md, curated) — the complete verification package |
P1 |
theory/ |
Layer-2 synthesis: consolidated sector expositions citing claim IDs at registered tiers | P1 |
verification/ |
Claim linter, verification scripts, tests | P1 |
codes/ |
Numerical codes by domain (vacuum, topology, gravity, flavor, cosmology) | P1 |
predictions/ |
Prediction ledger with input-freeze protocol | P1 |
negative-results/ |
Registry of failed branches and retracted claims | P1 |
reviews/ |
External review rounds and errata | P1 |
publish/website/ |
Website-bound curated content (generated from claims) | P2 |
publish/papers/ |
Paper manuscripts, one folder per paper | P2 |
archive/ |
Curated migration target for the legacy corpus (2024–2026) | P1 |
internal/ |
Local-only working area — never synced to GitHub | P0 |
Publication tiers P0/P1/P2 are defined in governance/publication-tiers.md.
# 1. Validate the claim ledger (schema, DAG, tier-monotonicity)
python verification/scripts/lint_claims.py
# 2. Read the master ledger
# CLAIMS.md — every claim, its tier, its falsifier, its open gates
# 3. Pick a claim and try to break it
# claims/<ID>/claim.md — statement, scope, falsifier, reproduction commandSee REVIEWING.md for the recommended attack surface.
The 2024–2026 research record (≈440 math notes, solvers, run archives) lives in a
separate legacy repository (TECT2/Contents). It is being migrated into this
repository pull-based and re-validated, claim by claim — never bulk-copied.
Disposition of every legacy file is tracked in archive/MIGRATION-LEDGER.md.
Until a legacy evidence pointer is migrated, claim cards cite it with the prefix
legacy: and the claim cannot rise above tier T6 (see
governance/migration-plan.md).
All tracked files are English-only. Korean is reserved for the conversational layer with AI collaborators and never appears in tracked content.