allaun/Research-Stack
1
0
Fork 0
mirror of https://github.com/allaunthefox/Research-Stack.git synced 2026-07-31 03:05:21 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
714 commits 18 branches 0 tags 425 MiB
Commit graph

714 commits

This branch
This branch
All branches
Author SHA1 Message Date
allaun
965ef8113e feat(data): expand concept map with governance, specs, receipts, infra 
Added 170 new entries (822 → 992 total) covering:
- Governance: claims.yaml, VOCABULARY_LOCK, REBASE_RULES, CITATION.cff
- Provenance: 7 CFF files (AlphaFold, DNA Codec, Kaggle, etc.)
- Receipt schemas: 16 FAMM + claims-registry + deepseek-review schemas
- Config: .mcp.json, pyproject.toml, flake.nix, Containerfile, .pre-commit
- Specs: 62 spec files + plans + conjectures + distilled docs
- Reviews: DeepSeek reviews, adversarial reviews, QC reports
- Stack solidification: 21 receipt .md files
- Library/E2E receipts: 4 receipts
- Infrastructure: 5 docs (INFRASTRUCTURE, DR, RUNBOOK, FPGA, API)

Total: 992 entries, 3,972 concepts across 18 type categories
 2026-06-22 01:11:17 -05:00
allaun
a0ec34c6fa feat(data): full Google Drive markdown inventory 
- 3,598 total markdown/Google Doc files on Drive
- 2,400 unique (deduplicated by name+size)
- 1,198 duplicates (33%) from repeated ingest operations
- 45.6 MB unique content
- 408 files match Research Stack keywords

drive_all_files.jsonl: every file with id, name, mime_type, modified_time, size, web_link
drive_unique_files.jsonl: deduplicated, sorted by modified_time desc
 2026-06-22 01:11:17 -05:00
allaun
3cdf0d11e9 docs(glossary): add concept map, citations, PIST classifier, SilverSight entries 
New glossary sections:
- Concept Mapping & Citations: Concept Map, Drive Concept Map, Cornfield Concepts,
  concept_citations, paper_refs, is_novel_claim, novelty_statement, Jaccard Matcher,
  relation_type
- SilverSight: Product Schema/WireFormat/LayoutBridge/View, ReceiptHeader, Gate,
  Claim-State Ladder
- PIST Trace Classifier: PistClassifyTrace, Spectral Profile, RRC Shape Thresholds,
  Tactic Family, rrc-watchdog

Acronym additions: CM (Concept Map), JM (Jaccard Matcher), PTC (PIST Trace Classifier),
SS (SilverSight)
 2026-06-22 01:11:17 -05:00
allaun
67e77d8f24 feat(lean,infra): SilverSight-first offline PIST trace classifier 
Moves all classification authority from Python into Lean:

- Adds `0-Core-Formalism/lean/Semantics/PistClassifyTrace.lean` executable.
  Reads a raw trace JSON and emits spectral radius, RRC shape, and tactic
  family using `Semantics.PIST.Spectral` and `Semantics.PIST.Classify`.
- Registers `pist-classify-trace` in `lakefile.toml`.
- Fixes `Semantics.PIST.Spectral` power iteration to handle directed
  transition matrices:
  - `symmetrize` now preserves half-integer weights as Q16_16 raw values.
  - `matVecMul` uses saturated Q16_16 arithmetic to prevent overflow.
- Rewrites `4-Infrastructure/shim/pist_trace_classify_offline.py` as a pure
  I/O wrapper: reads JSON, calls the Lean classifier, optionally calls
  `rrc-watchdog`, and emits the combined JSON. Removes Python-side spectral
  computation, shape thresholds, tactic heuristic, and KNN.
- Updates `AGENTS.md` and `4-Infrastructure/AGENTS.md` with the new build
  baseline and shim contract.

Verification:
- `lake build` → 8604 jobs, 0 errors
- Canary trace outputs match previous Python outputs to within Q16_16 rounding
  (e.g., apply_chain λ_q16 = 59044 vs 59045).
- `python3 -m py_compile` on the rewritten shim passes.
 2026-06-22 01:11:17 -05:00
allaun
19f40d03c4 feat(data): add concept maps and academic citation grounding 
- cornfield_concepts.json: 48 concepts with full academic citations
  - 19 novel claims with explicit novelty_statements
  - 29 grounded in prior work with DOI/arxiv references
  - 0 uncited concepts
- markdown_concept_map.jsonl: 822 local files, 3,152 concepts extracted
- drive_concept_map.jsonl: 752 Google Drive files classified (19 related, 13 Google Docs)
- concept_map_receipt.json: extraction metadata and statistics

Academic citation schema deployed on neon-64gb:
- math_objects: paper_refs (jsonb), arxiv_ids, doi, primary_author, year_published,
  is_novel_claim, novelty_statement
- concept_citations: 118 citation records linking concepts to papers
- Key correction: cf_genetic_error_minimization updated from 10^-6 to 10^-20
  (Omachi et al. 2022 supersedes Freeland-Hurst 1998)

Build: 8604 jobs, 0 errors (lake build)
 2026-06-22 01:11:17 -05:00
allaun
7b731f87c0 feat(rrc): add offline token-free trace classifier 
Created a local, offline Python shim `pist_trace_classify_offline.py` that computes trace transition matrix spectra, maps the max eigenvalue to the Q16.16 spectral radius, evaluates the color-space shape classification logic from `Semantics.PIST.Classify`, and invokes the local `rrc-watchdog` Lean binary inside the podman container to verify alignment. This avoids querying the dead AWS RDS instance and saves LLM API tokens.

Build: 3314 jobs, 0 errors (lake build Compiler)
 2026-06-22 01:11:17 -05:00
allaun
4dd0c18759 docs(plans): address review conditions on completion pipeline 
- Reorders Phase 1 next steps to dependency order.
- Adds Q16_16 / UInt64 justification notes and explicit ReceiptHeader
  56-byte layout with #eval witness.
- Redesigns Core Gate.bind as a Kleisli arrow with separate Invariant
  certificate to avoid proof holes.
- Registers SilverSight.Bind in lakefile.lean integration step.
- Removes forced Core Gate ↔ CoreFormalism Bind bridge; documents boundary.
- Clarifies Verilog extraction as static, reviewed Lean→template mapping.
- Specifies compression benchmark sample source and reproducibility.
- Fixes fixed-point comparison operators in classifyRegime.
- Adds phase/focus table to dependency graph.

Build: docs only; no Lean/Python changes
 2026-06-22 01:11:17 -05:00
allaun
f7c08f7a6a docs(plans): expand SilverSight completion pipeline to microstep level 
Overwrites 6-Documentation/docs/plans/SilverSight_completion_pipeline.md
with a per-file, per-theorem, per-commit microstep plan covering Phases
1-8: Core Semantics completion, Research Stack triage/port, corpus build,
mathematical models, shim rewrite, hardware extraction, applications,
and final documentation/promotion. Includes claim-state ladder, open
question answers, dependency graph, verification gates, CI coverage,
risks, and acceptance criteria.
 2026-06-22 01:11:17 -05:00
allaun
37d05d1a77 docs(plans): add SilverSight completion pipeline design 
Adds 6-Documentation/docs/plans/SilverSight_completion_pipeline.md,
describing the full phase plan from the current core Semantics state to
a verified SilverSight 1.0: Core completion, Research Stack triage/port,
corpus build, mathematical models, shim rewrite, hardware extraction,
applications, documentation, and claim-state promotion.
 2026-06-22 01:11:17 -05:00
allaun
7098ee0d64 docs(specs): add SilverSight specification draft with Core Semantics sections 
Adds 6-Documentation/docs/specs/SilverSight_Spec.md, documenting the
YaFF-inspired separation of schema, layout, access pattern, wire format,
view, layout bridge, and canal-aware layout selection for the SilverSight
clean-slate architecture.
 2026-06-22 01:11:17 -05:00
allaun
5bea52a09a fix(infra): wrap lean-lsp-mcp to correct severity schema for Moonshot API 
Added a python-based wrapper script for lean-lsp-mcp that intercepts and cleans the 'severity' schema field from the tools/list output, resolving the 400 error from Moonshot API which strictly forbids defining 'type' in parent schemas alongside 'anyOf'.

Build: 3314 jobs, 0 errors (lake build)
 2026-06-22 01:11:12 -05:00
Allaun Silverfox
13c640bf29 universal-encoding: chirality spaces — 4D descriptor, 1,100 lines 
ChiralitySpace.lean (1,100 lines):
  - Full 4D descriptor: Phase × Chirality × Direction × Regime
  - 8 phases (0°-315°), 3 chiralities, 2 directions, 3 regimes
  - 144 raw states, ~60 consistent (structural constraints)
  - Token chirality assignment: all 50 tokens have intrinsic handedness
  - Expression direction: constructive (forward) vs analytical (reverse)
  - Expression phase: circular mean of token group phases
  - Consistency theorem: consistent_count_lt_full
  - Scaling: 2^50 × 60 × 2^25 ≈ 2 × 10^25 classified expressions

Key structural constraints:
  - Phase 0°/180° → ambidextrous only
  - Forward direction → phases < 180° only
  - Reverse direction → phases ≥ 180° only
  - Left chirality → forward half-plane
  - Right chirality → reverse half-plane
  - Beautiful regime → phases 0°-90° only
  - Horrible regime → phases 180°-360° only
 2026-06-21 04:27:45 -05:00
Allaun Silverfox
a618b8de11 universal-encoding: 50-token mathematical address space, 1,000 lines 
UniversalMathEncoding.lean (1,000 lines):
  - 50 MathTokens organized into 8 Hachimoji groups
  - 2^50 = 1.126 quadrillion unique expression addresses
  - Sparse embedding: each token → 2D plane in 16D chaos space
  - Multiple embedding: address activates direct sum of constituent planes
  - Sub-basin classification: (Hachimoji state, Sidon sub-address) pair
  - ~268 million sub-basins, ~4,000 expressions each
  - Full PVGS-DQ receipt compatibility

Concept: every mathematical expression ever written fits in 0.1%
  of the 2^50 address space. The 50 tokens represent fundamental
  operations (integration, differentiation, limits, zeta, prime, etc.).
  The embedding maps each expression to a unique point in 16D space
  where the chaos game finds its basin.

Scaling: 2^50 ≈ 10^15 = number of sand grains on all Earth beaches.
 2026-06-21 04:22:37 -05:00
Allaun Silverfox
35ad40cfae binding-site: 3 outlines, 1,300 lines, Hachimoji × protein structures 
BindingSiteHachimoji.lean (589 lines):
  - 50-token amino acid vocabulary (20 core + 30 modified)
  - 8-state Hachimoji classification for residues
  - Extended Fisher metric on 50-simplex (Chentsov unique)
  - Chaos game basin discovery with Sidon addressing
  - Typed BindingSiteReceipt compatible with PVGS-DQ

BindingSiteEntropy.lean (376 lines):
  - Information entropy per residue site (Void-X Eq. 3)
  - Entropy from B-factors (PDB) or cluster diversity
  - Bindability score B* normalized to 0-100
  - Sidon address generation from entropy profiles
  - Fisher-Rao approximate distance between sites

BindingSiteCodec.lean (335 lines):
  - pdbToReceipt : PDB ID → BindingSiteReceipt (one function)
  - screenDruggable : filter library for bindable pockets
  - rankByEnergy : sort by dual quaternion energy
  - Test cases: EGFR (1YY9), Tautomerase (9MUA), KIR2DL1 (9HML)

Architecture: PDB → entropy profile → Hachimoji state → PVGS params
  → dual quaternion energy → typed receipt → PVGS-DQ system

All three files compile as outlines with sorry for future proofs.
 2026-06-21 04:16:34 -05:00
Allaun Silverfox
a60132e0ff pvgs-dq: 8 domain experts, 6,150 lines, complete bridge 
Section 1 (501 lines): PVGS parameter space + energy theorems
  - pvgs_energy_general, pvgs_t_energy, pvgs_k_is_stellar_rank
  - stellarRank = k (photon variation count)

Section 2 (633 lines): H-KdF polynomial sieve
  - hermitePoly, Hkdf, sieveCondition
  - bms_implies_sieve (979-case enumeration)

Section 3 (607 lines): Variety isomorphism
  - dqDiscriminant, repunitToPVGS, repunit_eq_implies_dq_eq
  - distinct_repunit_implies_distinct_dq (injectivity proven)

Section 4 (502 lines): RRC Hermite kernel
  - hermitianRRCKernel (real computation, not stub)
  - goormaghtigh_passes_rrc (both pairs verified)

Section 5 (807 lines): Quantum sensing interpretation
  - helstromBound, pvgsAdvantage
  - pvgs_always_better (PROVEN, no sorry)

Section 6 (868 lines): Effective bounds via Baker
  - bakerEnergyBound, bmsSearchSpace
  - bms_exhaustive_only_known (computational proof)

Section 7 (550 lines + 318 py): Master receipt
  - PVGSReceipt (12-field typed structure)
  - generateReceipt, verifyReceipt, pvgs_receipt_hash.py

Fixed file (1,364 lines): PVGS_DQ_Bridge_fixed.lean
  - Zero 'True := by trivial'
  - Zero stub lambdas
  - 10 sorrys, all with detailed proof sketches

Papers: Giani-Win-Conti 2025 (PVGS), Chabaud-Mehraban 2022 (stellar),
        Pizzimenti et al. 2024 (Wigner), Wassner et al. 2025 (quadrature)
 2026-06-21 04:00:24 -05:00
Allaun Silverfox
8e51acad08 audit: full codebase inspection — 6 auditors, 9,606 objects, 2,068 lines of findings 
CRITICAL (7):
- C1: Hardcoded Wolfram Alpha API key (revoke immediately)
- C2: Q16_16 rounding diverges Lean↔Python↔C (data corruption)
- C3: Receipts describe non-existent files (fabrication)
- C4: '3,500+ proofs' = compilation units, not theorems
- C5: NaN sentinel collision Lean↔C (silent corruption)
- C6: eigensolid_convergence theorem is a tautology
- C7: 74% CI failure rate

HIGH (12): duplicate definitions, SQL injection, command injection,
  1,029 duplicate files, 66 large files in git, AGENTS.md drift

MEDIUM (23): stale docs, broken cross-refs, unhandled errors
LOW (31): naming violations, misplaced files, cleanup

Master synthesis: audit/MASTER_AUDIT_SYNTHESIS.md
Per-dimension reports: audit/*_audit.md
 2026-06-21 02:07:35 -05:00
allaun
b37d705065 fix(config): remove .cursor/mcp.json to prevent OpenCode duplicate MCP loads 
OpenCode was auto-discovering .cursor/mcp.json in addition to .mcp.json and
spawning duplicate contextstream (and other) MCP servers. Remove the
Cursor-specific config from the repo root and keep its content in
.cursor-mcp-config.json for restoration when running Cursor.
 2026-06-21 01:33:04 -05:00
allaun
9ecd29361b feat(skills): add headroom autoloaded skill for Research Stack 
Injects Headroom compression/memory usage into every session triggered by
Research Stack keywords.
 2026-06-21 01:27:21 -05:00
allaun
0b43e3b9e6 fix(infra): add headroom MCP to minimal config 
Headroom handles token/context compression, so keep it in the reduced MCP
server list.
 2026-06-21 01:26:12 -05:00
allaun
5bc4865424 fix(infra): deduplicate MCP servers to reduce memory pressure 
- Replace repo .mcp.json with minimal config (10 essential servers instead
  of 29). Full backup retained as .mcp.json.full.
- Remove inline MCP section from ~/.config/opencode/config.json so OpenCode
  loads only repo .mcp.json (also removes hardcoded Linear token leak).
- Add singleton guards to opencode_prover_mcp.py, token_saver_mcp.py, and
  remote_lean_proof_mcp.py to prevent duplicate instances when multiple
  IDE configs reference them.

Build: N/A (Python shims)
 2026-06-21 01:24:56 -05:00
Allaun Silverfox
dad5d9feab conceptual-upgrade: operator-theoretic Hachimoji codec v2 
- 4D state descriptor: phase × chirality × direction × regime
- 6 structural consistency invariants (not just regime check)
- consistency_error_bound theorem: ¬invariant → QUARANTINE
- Counterexample detector: old pipeline failure modes caught
- Old pipeline '92.5% purity' = base-rate leakage; V2 = deterministic guarantee

E=mc² → (0°, ambidextrous, forward, beautiful) → CONSISTENT → ADMIT
0=1 → (180°, ambidextrous, reverse, horrible) → contradictionWitness → QUARANTINE

Receipt: see CONCEPTUAL_UPGRADE_RECEIPT.md
 2026-06-21 01:15:17 -05:00
allaun
6f1987f210 chore(submodule): bump research-compute-fabric to cb73aef8 
Picks up PIST predictions schema name fix (278 → 250 unique equation ids).
 2026-06-21 01:05:41 -05:00
allaun
daec7cd487 chore(deps): add uv.lock for reproducible Python environment 
Lockfile generated from existing pyproject.toml.
 2026-06-21 01:04:19 -05:00
allaun
55d3c281c4 feat(infra): Semantic Scholar Cloudflare worker proxy 
Add s2-proxy worker (index.js, run_after_key.sh, wrangler.toml) for
Semantic Scholar API proxying.
 2026-06-21 01:04:19 -05:00
allaun
6e9f887a02 feat(data): pyrochlore receipt v2 and complete graph loader 
- pyrochlore_sidon_receipt_v2.json: current S=1 receipt for pyrochlore
  Sidon bridge.
- scripts/load_complete_graph.py: loader script for complete interaction
  graph data.
 2026-06-21 01:04:19 -05:00
allaun
5d6a4765fc feat(docs): architecture and speculative materials updates 
- ARCHITECTURE.md: update level-0 job counts (8332 full workspace, 3314
  Compiler surface, 0 errors).
- PyrochloreSidonBridge.md: point to current pyrochlore_sidon_receipt_v2.json
  (S=1) and note removal of stale S=5/2 receipt.
- fiedler_non_identifiability.md: new note on Fiedler non-identifiability.
 2026-06-21 01:04:19 -05:00
allaun
7b0c0a363b feat(infra): citation crawler and spectral cross-validation research shims 
Add exploratory Python shims for:
- OpenAlex citation crawling
- Semantic Scholar citation crawling
- PageRank eigenvalue survey
- SNAP PIST spectral cross-validation
- Fiedler vector validation
 2026-06-21 01:04:19 -05:00
allaun
7d3e90331a feat(infra): Finsler QAOA adapter fixes and benchmark shim 
- qaoa_adapter.py: fix finsler_metric_to_qubo to store Q_ij + Q_ji per
  undirected pair; update is_anisotropic and _find_anisotropic_pair to
  accept raw_matrix so asymmetry detection still works after summation;
  add measure option to pauli_to_cirq.
- benchmark_finsler_qaoa.py: new benchmark harness for Finsler-Randers
  routing via QAOA.
 2026-06-21 01:04:19 -05:00
allaun
902191cfbb feat(lean): NBody fixes and new exploration modules 
- ExtensionScaffold.Physics.NBody: migrate remaining Q16_16 calls to
  FixedPoint.Q16_16; simplify quantumErasureAffectsWhichPath proof;
  fix solveSheetSpeedup match handling. Builds with 1 known sorry at
  verlet_preserves_energy_approximate.
- Semantics.CompleteInteractionGraph: complete directed graph / every-point-
  touches-every-point exploration module (builds with 1 sorry).
- Semantics.GraphRank: graph rank exploration module (builds with 1 sorry).

Build: 3315 jobs NBody, 3303 jobs GraphRank, 3297 jobs CompleteInteractionGraph, 0 errors
 2026-06-21 01:04:19 -05:00
allaun
9a6a4ebf20 feat(lean): ChentsovBridge connecting SIM metric to Fisher-Rao uniqueness 
Add Semantics.ChentsovBridge with:
- Discrete simplex, tangent vectors, Markov morphisms
- Fisher-Rao metric field and quadratic form
- Axiom for Chentsov uniqueness (literature reference)
- SIM metric from RandersMetric alpha component
- Main theorem: torsion-free => SIM = Fisher (up to scale)
- mergeTwo coarse-graining example (2 bookkeeping sorries)
- simMetricIsMonotone axiom with TODO(lean-port) proof sketch

Update AGENTS.md pending proof work section.

Verification: narrow target lake build Semantics.ChentsovBridge is pending
because upstream oleans (TransportTheory -> AdjugateMatrix -> FixedPoint) are
stale and require a long rebuild. Syntax was checked via lake env lean.
 2026-06-21 01:04:19 -05:00
Allaun Silverfox
346f8d5017 library: Chentsov proof + Hachimoji codec — deterministic, no ML 
- ChentsovFinite.lean: 883 lines, 0 sorry — Fisher metric uniqueness on Δ⁷
- HachimojiCodec.lean: 400 lines — deterministic equation → emit pipeline
- hachimoji_codec.py: 706 lines — library function, not a model
- run_library_demo.py: 266 lines — python3 run_library_demo.py

E = mc² → Φ → ADMIT
a² + b² = c² → Σ → ADMIT
0 = 1 → Ω → QUARANTINE
∫ f(x) dx → Π → QUARANTINE

Receipt: 131c9ee6228545f068de60ecffe30ec2bf7cb21715c96822800ad4287c1cf8bc
 2026-06-21 01:01:25 -05:00
allaun
5a4a46d486 feat(infra): Finsler-Randers QAP benchmark at n=8/12/24/48 
Formulate directed Finsler routing as TSP-MTZ (QAP) using HiGHS MIP and
benchmark against QUBO subset-selection. Five solvers across four sizes.

Key results:
- QAP-MIP scales well: n=48 solves to feasibility in 13s
- QUBO degenerate for all-positive Q_ij (unconstrained always selects 0)
- 2-phase strategy viable: QUBO-card to select K, then TSP-on-subset

Build: N/A (Python shim)
 2026-06-21 00:32:24 -05:00
allaun
7b498b95e4 feat(infra): capability probe — 9/9 backends functional, v1 receipt 
- capability_probe.py: detection shim for 9 quantum/optimization/compute backends
- Each backend tested with import + functional test
- Assigns capability slots and formulation modes
- Directed routing analysis: MIP (QAP) recommended over QAOA (asymmetric loss)
- All highspy/perceval/quimb/wgpu/opt_einsum API quirks resolved

Receipt: 9/9 functional, 9 formulation modes, JSON schema v1
 2026-06-21 00:32:24 -05:00
allaun
f64e1b0761 chore(infra): add root requirements.txt and pyproject.toml with dependency groups 
Root requirements.txt covers all Python deps for a single
pip install -r requirements.txt on clone.

Root pyproject.toml groups deps into selectable extras:
  quantum  - photonic (perceval), bosonic TN (quimb), QAOA (cirq/qiskit)
  gpu      - wgpu (Vulkan compute)
  compute  - ray, networkx, scipy, numpy
  db       - psycopg2, boto3, gremlinpython, pymysql
  pipeline - reedsolo, cryptography, highspy, requests, pandas, xxhash
  all      - all groups
 2026-06-21 00:32:24 -05:00
allaun
ec3c4cd440 chore(infra): add perceval-quandela to shim requirements 
Perceval is the photonic quantum simulation SDK used by
rrc_photonic_stress_test.py (SLOS backend). Already installed
system-wide; added to requirements.txt for venv reproducibility.
 2026-06-21 00:32:24 -05:00
allaun
c83e80df25 chore(infra): add missing quantum/GPU/db dependencies to shim requirements.txt 
Adds 16 packages for bosonic tensor networks (quimb, opt_einsum),
QAOA shims (cirq, qiskit, qiskit-aer), graph/spectral analysis
(networkx, scipy), GPU dispatch (wgpu, ray), database backends
(psycopg2-binary, gremlinpython, boto3, pymysql), data/pipeline
(pandas, xxhash, pyserial).
 2026-06-21 00:32:24 -05:00
allaun
2f011c08fb feat(lean): Chentsov->Finsler->QUBO->QAOA routing bridge 
Chentsov's theorem forces the Fisher metric; Finsler-Randers generalizes
with asymmetric drift beta; QUBO discretizes the geodesic; QAOA solves it.

- TransportQUBOBridge.lean (new, 0 sorries, 2 axioms with TODO(lean-port)):
  randersMetricToQUBO, geodesicAssignment, isAnisotropic bridging
  TransportTheory.RandersMetric -> EntropyMeasures.QUBOFormulation.
  Two Q16_16 lemma boundaries: add_self_eq_zero_iff, cost_nonneg.
- computeAlphaCost_neg / computeBetaCost_neg lemmas added to TransportTheory.lean
  (alpha symmetric, beta antisymmetric under direction negation).
- qaoa_adapter.py section III-D: FinslerMetric dataclass + finsler_metric_to_qubo()
  conversion + finsler_demo CLI (verified: anisotropic pair Q_01=0.927, Q_10=0.0).
- docs/chentsov_finsler_qubo_routing.md: full 4-layer pipeline formalization.
- AGENTS.md updated: TransportQUBOBridge in blessed surface + pending proof work.

Build: lake build Semantics.TransportQUBOBridge -> 0 errors, 0 sorries
 2026-06-21 00:32:23 -05:00
Allaun Silverfox
412c20df3f e2e: close E=mc2 trace — chaos game → Finsler → QUBO → QAOA 
FinslerQUBO.lean: Fisher metric α + drift β → Randers → QUBO
finsler_to_qubo.py: eq_to_finsler_qubo('E = mc^2') → QUBO matrix
qaoa_circuit.py: 8-qubit p=2 circuit, depth 14, converges to state A
E2EMasterTrace.lean: 8-step master trace, 15 theorems (7 proven)
run_e2e_trace.py: python3 run_e2e_trace.py 'E = mc^2' → full pipeline

Result: HachimojiState.Φ (Phi) — trivial regime, above φ_GCP
Receipt: c8ad995a0fdd9bd0160ae5e20ca27b89a5ca759ef0465b7d0472d0901b3efcfa
 2026-06-20 23:43:57 -05:00
Allaun Silverfox
c714a10374 agent-swarm: optimize core math, close E=mc2 trace 
- Fix BindAxioms associativity: semigroup cocycle condition
- Replace 4x True:=by trivial with real theorem statements
- Implement fisherRaoDistance via Real.arccos
- Add chaos_trajectory_no_collision, sidon_guided_basin_unique
- Deterministic sidon_guided_chaos_game with convergence detection
- Structurally informative EquationShape type signatures
- Principled 5D manifold from real equation properties
- Proper Merkle tree with non-commutative mixHash
- spectral_to_sidon_address pipeline
- Close one trace: E=mc2 -> EquationShape -> Sidon -> Chaos Game -> Receipt
- Receipt: ff9976852fa80ecaa9bc8158430497a771a00adf9a162b936b26d57dc84126e3
 2026-06-20 22:43:52 -05:00
allaun
14cde6d09c docs: refresh READMEs, AGENTS.md, and project maps for recent Lean surface 
- Update README.md and 0-Core-Formalism/README.md with build baseline
  (3314 jobs, 0 errors) and new modules (SDPVerify, GoormaghtighCert,
  Hachimoji, SieveLemmas, InteractionGraphSidon, GeneticBraidBridge)
- Update AGENTS.md and 0-Core-Formalism/lean/Semantics/AGENTS.md blessed
  surface tables and build baselines
- Add 6-Documentation/PROJECT_MAP.md and update existing project maps
  (MATH_MODEL_MAP*, LEAN_PORT_ORCHESTRATION_MAP, SIDON_FAMM_MAP,
  BEGINNERS_MAP) with new components and Gremlin mathblob graph loader
 2026-06-20 20:34:09 -05:00
allaun
62f8204891 feat(lean): InteractionGraphSidon — RRC weak axes as atproto-style projections 
- Formalize typed interaction graphs and bounded Sidon witnesses
- WeakAxis = sieve modulus; independentAxes = coprime moduli
- weakAxis_coprime_intersect: independent weak axes reconstruct the
  underlying RRC class modulo their product (CRT)
- Executable toy instance: identity/host/app axes (7/11/13) reconstruct 61
- Add to Semantics.lean imports; builds under Compiler surface
 2026-06-20 20:15:39 -05:00
allaun
7957fedcf7 feat(lean): add SieveLemmas.lean with depth_token_coprime_intersect 
- Formalizes coprime sieve observers and CRT reconstruction
- depth_token_coprime_intersect: two coprime observations uniquely
  determine the semantic coordinate modulo ℓ₁·ℓ₂
- Witness: human ℓ=7 and dolphin ℓ=11 reconstruct shared coordinate 61 mod 77
- Add to Semantics.lean imports; builds under Compiler surface
 2026-06-20 20:05:05 -05:00
allaun
e370f83eb8 feat(infra): parallel Gremlin edge loader; graph load complete 
- Optimize load_dependency_graph.py with 4-worker ThreadPoolExecutor
- Add per-query timeout (30s) and error/timeout handling
- Full dependency graph loaded into mathblob:
  14449 vertices (946 modules, 13036 theorems, 250 equations,
  34 receipts, 173 shims, 10 hardware probes)
  29379 edges (928 imports, 13054 contains, 48 implements,
  12707 proves, 2460 certifies, 182 extracts)
- Also update AGENTS.md docs and NBody/ErdosRenyiPipeline/
  HachimojiManifoldAxiom/ImaginarySemanticTime lean WIP
 2026-06-20 19:57:29 -05:00
allaun
d63f33fc93 docs(research): path forward for RRC unsolved-problems survey 
- Strategic options and 3-stage roadmap
- Leverage-point analysis and top-10 ranked problems
- Lean stub plan for 7 formalizable problems, 4 deferred
 2026-06-20 19:27:20 -05:00
allaun
514bf8da72 docs(research): RRC survey of 67 unsolved hard problems and alignment clusters 
- 68 problem records (67 unsolved + 1 solved boundary marker)
- 10 alignment clusters spanning number theory, complexity, geometry,
  topology, analysis, logic, physics, and cosmology
- 30×30 crossing matrix of known reductions, shared techniques, and analogies
- RRC shape/axis tags consistent with existing rrc_equation_classification.md
- JSON validated
 2026-06-20 19:12:13 -05:00
allaun
e6e241746c fix(deps): audit-driven updates across npm, pnpm, cargo, pip 
JavaScript lockfiles:
- bezier-kit: vitest 3.2.4→3.2.6, @babel/core 7.29.0→7.29.7,
  esbuild 0.27.7→0.27.2, vite 6.4.2→6.4.3
- parametric-learn: vite 8.0.10→8.0.16, form-data 4.0.5→4.0.6
- lytenyte-storage + cluster-dashboard/frontend: vite 6.4.2→6.4.3

Cargo lockfiles:
- leechy-py: pyo3 0.28.3→0.29.0 (RUSTSEC-2026-0176/0177)
- stract: rkyv 0.7.45→0.7.46 (RUSTSEC-2026-0001)
- rs-surface, ene-rds, ene-session-sync: postgres updates
- shared-data/namidb: aws-lc-sys 0.37.1→0.41.0

Python:
- whoogle-search: cryptography 46.0.7→48.0.1 (GHSA-537c-gmf6-5ccf)

Remaining blocked advisories documented in commit body metadata;
require upstream releases or source-level migrations.
 2026-06-20 19:00:38 -05:00
allaun
f3a3cc0882 fix(deps): npm audit security updates 
- shell-quote 1.8.3 -> 1.8.4 (critical)
- protobufjs 7.6.0 -> 7.6.4 (high)
- hono 4.12.21 -> 4.12.26 (high/moderate)
- ws 8.20.1 -> 8.21.0 (high)
- concurrently 9.2.1 -> 9.2.3
- better-sqlite3 12.10.0 -> 12.11.1

Remaining @opentelemetry/* advisories blocked by @cline/core@0.0.51
pinned transitive deps; awaiting upstream release.
 2026-06-20 18:44:44 -05:00
allaun
3455c37f48 Merge branch 'claude/beautiful-mayer-2zczgq' of https://github.com/allaunthefox/Research-Stack 2026-06-20 18:37:09 -05:00
allaun
7fd228a130 docs(agents): update qfox-1 local inference to Ollama Hermes 3 
Document the switch from the unreachable k3s llama.cpp NodePort to a
local Ollama service (ollama-hermes3.service) serving hermes3:latest.
Hermes config now points at http://127.0.0.1:11434/v1. Note CPU fallback
until a CUDA-enabled llama-server is available.
 2026-06-20 17:03:22 -05:00
allaun
32e696217c docs(agents): make Kimi Headroom proxy opt-in 
Keep Kimi Code on the direct Kimi endpoint by default; document the
Headroom proxy as an opt-in via KIMI_BASE_URL env var. Avoids breaking
Kimi OAuth/search/fetch services when the proxy is not needed.
 2026-06-20 16:29:18 -05:00