From bb3cfd046431ef1efa77adbe77f28031f9a9cae4 Mon Sep 17 00:00:00 2001 From: axiomlogicnexus Date: Wed, 27 May 2026 22:06:18 +0200 Subject: [PATCH] Reconcile yakupbilen research benchmark lane --- ...RCH_HIERARCHY_RECONCILIATION_2026-05-27.md | 159 ++++++++++++++++++ ...SE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md | 9 +- ...NED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md | 2 +- ..._AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md | 2 +- .../HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md | 2 +- docs/REPO_LICENSE_TRACKING.md | 16 +- ...epo_portfolio_unified_operational_v6_3.csv | 2 +- docs/repo_portfolio_unified_phase_g_v6_3.csv | 2 +- ...po_portfolio_unified_source_audit_v6_3.csv | 4 +- .../HyperTwist/FEATURE_REGISTRY.md | 1 + 10 files changed, 188 insertions(+), 11 deletions(-) create mode 100644 docs/HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md diff --git a/docs/HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md b/docs/HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md new file mode 100644 index 0000000..6509025 --- /dev/null +++ b/docs/HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md @@ -0,0 +1,159 @@ +# HyperTwist Learned Heuristic Search And Offline Solver Research Hierarchy Reconciliation - 2026-05-27 + +## Status + +This document is the source-backed reconciliation note for the retained +`yakupbilen/drl-rubiks-cube` row against the current HyperTwist stack. + +It answers one narrow question: + +- does the current `yakupbilen/drl-rubiks-cube` source justify reopening any + live HyperTwist perception, reconstruction, solver, or training owner lane + +Result: + +- no product-code reopening is required +- yes selective authority backfill is required because older row-source surfaces + still over-describe the repo as a vision/perception donor instead of the + narrower research benchmark it actually is + +## Source basis + +- `C:\HyperTwist\docs\HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md` +- `C:\HyperTwist\docs\REPO_LICENSE_TRACKING.md` +- `C:\HyperTwist\docs\HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md` +- `C:\HyperTwist\docs\HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md` +- `C:\HyperTwist\docs\HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md` +- `C:\HyperTwist\docs\v6_5_deep_manual_pack\HyperTwist\FEATURE_REGISTRY.md` +- `C:\Workspaces\HyperTwist\repos.manifest.json` +- `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\README.md` +- `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\cube\cube.py` +- `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\search\search.py` +- `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\train\train_utils.py` +- `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\networks\modelpaper.py` +- current first-party owner surfaces under: + - `C:\HyperTwist\UnrealHyperTwist\Source\UnrealHyperTwist` + - `C:\HyperTwist\docs\ROADMAP.md` + - `C:\HyperTwist\docs\ops\HYPERTWIST_IMPLEMENTATION_PHASE_1_KICKOFF.md` + +## Slice-local verdict + +| Slice | Primary authority | Secondary value kept | Route | Width | Result | +| --- | --- | --- | --- | --- | --- | +| Live classic-cube calibration, ordered face observation, webcam shell, committed-face reconstruction, and bounded browser recognition shells | first-party landed `qbr` / `rubix-cube-solver` packets | none required from `yakupbilen` | existing landed permissive route | `F2` | unchanged live implementation owner | +| Live bounded classic-cube explanation, recommendation, correction, and resolution seams | first-party landed `Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S` seams | none required from `yakupbilen` | existing landed permissive route | `F2` | unchanged live implementation owner | +| Compact correctness and heuristic-search oracle | `brownan/Rubiks-Cube-Solver` plus `efrantar/rob-twophase` as comparator | `yakupbilen` remains adjacent research value only | existing retained non-live route | `F2` | unchanged retained oracle order | +| Learned heuristic search, ADI state generation, batched neural search experimentation, and small value-network research | `yakupbilen/drl-rubiks-cube` | adjacent live recognition and retained correctness owners remain stronger for product use | `R4` now, `R1` only if a future first-party ML research lane is intentionally opened | `F2` | permissive research benchmark authority kept explicit | + +## Why no product-code reopening is required + +### `yakupbilen/drl-rubiks-cube` + +`yakupbilen` still has retained value, but it is narrow and research-oriented: + +- concise `54`-sticker state encoding with direct move-transition tables +- anti-cancellation move-sequence generation +- `ADI` state-generation and weighted target-preparation loops +- batched neural-evaluated `A*` search experiments +- small value-network research architecture + +Those strengths are explicit in the current source: + +- `cube.py` is centered on direct sticker-state transitions, scramble + generation, inverse-solution recovery, and sequence cleanup +- `train_utils.py` is centered on `ADI` training-state generation and target + construction rather than product shell behavior +- `search.py` is centered on batched node expansion and network-guided search +- `modelpaper.py` is a compact offline value model, not a deployable product + runtime + +That does not make the repo a meaningful live camera/perception owner. + +### Current live recognition and correction owners + +HyperTwist already owns the live physical-cube lane elsewhere: + +- `qbr` owns the landed calibration, ordered face observation, and webcam-shell + surface +- `rubix-cube-solver` owns the landed committed-face reconstruction and bounded + browser recognition shell +- current first-party correction and resolution shells are already live above + those seams + +`yakupbilen` does not reopen any of those owners because its strongest value is +offline research logic, not camera/calibration/runtime shell ownership. + +### Current retained solver-oracle owners + +HyperTwist also already has stronger retained solver-oracle anchors: + +- `brownan` remains the compact correctness and heuristic-table oracle +- `efrantar` remains the metric, pruning, robot-execution, and performance + comparator + +`yakupbilen` stays distinct as a learned-search and `ADI` experimentation +benchmark, not as the primary correctness or compact solver oracle. + +## Selective backfill result + +### Earlier packet reopened + +- `C:\HyperTwist\docs\HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md` + +### Reason + +The original `0R-E` call was directionally correct but later row-source layers +still blurred the repo into a perception/camera donor story. + +This pass makes the exact retained slice explicit: + +- `yakupbilen` is not a live recognition or camera-shell owner +- `yakupbilen` is not a current product donor despite permissive licensing +- `yakupbilen` remains the retained benchmark only for learned search, `ADI` + generation, and offline experimentation + +### Change class + +- no repo-local extraction changed +- no product-code change occurred +- no license-status change occurred +- row-source and dependent authority wording were sharpened + +## Propagation result + +### Feature registry + +Updated. + +The retained learned-search research slice is now explicit instead of living +only in packet and tracker prose. + +### Cross-lane doctrine + +No doctrine wording change was needed. + +This pass applied the existing `A/R/F` hierarchy; it did not change it. + +### Skillization + +No doctrine or `SKILLS.md` change was needed. + +This pass did not create or reopen any command/skill surface. + +### Legal evidence + +No license-status or custody change occurred. + +The manifest and tracker were sharpened instead. + +## Operational answer + +After this pass: + +- do not reopen live perception, reconstruction, or correction code +- do not reinterpret `yakupbilen` as a camera/perception donor +- do not promote it above `brownan` or `efrantar` for correctness/performance + oracle ownership +- if HyperTwist later opens a real first-party ML search or offline + experimentation lane, reopen only from the narrower learned-search/`ADI` + benchmark slice diff --git a/docs/HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md b/docs/HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md index 4c0cc2c..10a7605 100644 --- a/docs/HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md +++ b/docs/HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md @@ -741,8 +741,13 @@ Non-promotion decisions and rationale: Integration call: -- retain as MIT research benchmark material for learned heuristic search, offline experimentation, and future solver-evaluation work -- revisit only if HyperTwist explicitly opens a first-party ML search or experimentation lane +- retain as MIT research benchmark material for `54`-sticker transition + encoding, `ADI` state generation, batched neural search experimentation, and + future solver-evaluation work +- do not reinterpret the repo as a live camera/perception or product-shell + donor +- revisit only if HyperTwist explicitly opens a first-party ML search or + experimentation lane ## Phase 0R closure result diff --git a/docs/HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md b/docs/HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md index 89f1b4f..91ef29a 100644 --- a/docs/HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md +++ b/docs/HYPERTWIST_PHASE_1R_RETAINED_SET_CONTRACT_AND_HANDOFF_2026-05-13.md @@ -369,7 +369,7 @@ These rows stay retained, but not as active implementation backlog. - `AviKaufman/Rubix-cube-trainer` — primary all-rights-reserved guided-pedagogy benchmark - `alinen/cube` — no-license planner and task-segmentation benchmark - `ambisinister/blindsolve` — no-license BLD memo benchmark -- `yakupbilen/drl-rubiks-cube` — MIT learned-heuristic research benchmark +- `yakupbilen/drl-rubiks-cube` — MIT learned-heuristic, `ADI` state-generation, and offline solver-experimentation benchmark Contract: diff --git a/docs/HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md b/docs/HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md index ea9cfda..7c2714a 100644 --- a/docs/HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md +++ b/docs/HYPERTWIST_REPO_EVALUATION_RESET_AND_IMPLEMENTATION_SCHEDULE_2026-05-11.md @@ -262,7 +262,7 @@ Most important scheduling result: - `efrantar/rob-twophase` remains a secondary solver oracle for metric and performance comparison - `ShellPuppy/RCube` remains a large-`N` algorithm and parity benchmark - `AviKaufman/Rubix-cube-trainer`, `alinen/cube`, `ambisinister/blindsolve`, and `vwcwong/CubeSim` remain clean-room-later pedagogy, planner, memo, and history benchmarks -- `yakupbilen/drl-rubiks-cube` remains a permissive research benchmark for learned heuristic search and experimentation +- `yakupbilen/drl-rubiks-cube` remains a permissive research benchmark for learned heuristic search, `ADI` state generation, and offline solver experimentation - `aMonteSl/CodeXR`, `MathewKJ2048/Rubiks-cube-simulator`, and `brianpeiris/RiftSketch` are now discarded from the active retained implementation set Combined `Phase 0R` packet status: diff --git a/docs/HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md b/docs/HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md index adbac73..c24db70 100644 --- a/docs/HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md +++ b/docs/HYPERTWIST_REPO_STATE_BOARD_2026-05-11.md @@ -374,7 +374,7 @@ Count: `13` | `poliva/cubedex` | `no explicit license visible` | Locked Strategic Donor | `integrate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Phase 1R` plus the `2026-05-27` smartcube practice/review/SRS hierarchy clarification are closed. Retain only as restrictive comparison context for smartcube-aware practice-shell behavior, review/SRS UX, recognition-versus-execution timing presentation, and local stats/history comparison; do not treat it as a live owner or a default clean-room next row. | | `ShellPuppy/RCube` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a large-`N` centers/edges/parity benchmark and later clean-room research input, not as an active product donor. | | `vwcwong/CubeSim` | `GPL-3.0` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as a readable state/history and beginner-solver decomposition benchmark for later clean-room use. | -| `yakupbilen/drl-rubiks-cube` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation. | +| `yakupbilen/drl-rubiks-cube` | `MIT` | Reserve Bench | `future candidate` | `not_live_reference_or_discard_candidate` | `phase0r_reference_benchmark_or_discard_eval` | `Packet 0R-E` closed. Retain as an MIT research benchmark for `54`-sticker transition encoding, `ADI` state generation, batched neural `A*` search, and offline solver experimentation; do not treat it as a live camera/perception owner or a near-term product donor. | ## Implementation rule going forward diff --git a/docs/REPO_LICENSE_TRACKING.md b/docs/REPO_LICENSE_TRACKING.md index 6a26c5e..ca81d40 100644 --- a/docs/REPO_LICENSE_TRACKING.md +++ b/docs/REPO_LICENSE_TRACKING.md @@ -1768,11 +1768,23 @@ Source basis: - `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\train\train_utils.py` - `C:\Workspaces\HyperTwist\mirrors\permissive\yakupbilen\drl-rubiks-cube\networks\modelpaper.py` - `C:\HyperTwist\docs\HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md` +- `C:\HyperTwist\docs\HYPERTWIST_LEARNED_HEURISTIC_SEARCH_AND_OFFLINE_SOLVER_RESEARCH_HIERARCHY_RECONCILIATION_2026-05-27.md` Approved working posture: -- keep as a permissive research bench for learned heuristic search and offline experimentation -- do not let permissive status alone promote it above the committed training/perception/runtime core +- keep as a permissive research bench for `54`-sticker transition encoding, + `ADI` state generation, batched neural search experiments, and offline solver + experimentation +- do not treat it as a live camera, calibration, reconstruction, or browser + shell owner +- preserve `qbr` and `rubix-cube-solver` as the live recognition/reconstruction + owners +- preserve `brownan` and `efrantar` as the retained correctness/performance + solver-oracle owners +- do not let permissive status alone promote it into the committed + perception/training/runtime core +- reopen only if a first-party ML search or offline experimentation lane is + intentionally opened later ### `HactarCE/2x2x2x2-Scrambler` diff --git a/docs/repo_portfolio_unified_operational_v6_3.csv b/docs/repo_portfolio_unified_operational_v6_3.csv index 0fe3241..07ac76b 100644 --- a/docs/repo_portfolio_unified_operational_v6_3.csv +++ b/docs/repo_portfolio_unified_operational_v6_3.csv @@ -34,7 +34,7 @@ "11927","ambisinister/blindsolve","https://github.com/ambisinister/blindsolve","HyperTwist","28.0","117.0","132.0","B","clean-room BLD memo benchmark","benchmark","Clean-room BLD memo benchmark","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Benchmark-only retained value: narrow memo-practice behavior and BLD training flow worth preserving through a clean-room spec.","Keep only as a clean-room benchmark. Use it to shape a first-party memo-practice micro-mode; do not incorporate source.","Do not treat this repo as a merge candidate. Convert useful memo-practice ideas into a scrubbed Model A spec only.","Repurpose here means: derive first-party BLD memo-practice behavior through a clean-room process.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","clean-room benchmark review","Hidden value sits in narrow memo-practice flow and user-feedback behavior.","Which memo-practice behaviors are worth preserving through a scrubbed Model A handoff?","Inspect memo-practice flow, prompt cadence, scoring/feedback behavior, and narrow training-state transitions as clean-room benchmark material only.","Audit ambisinister/blindsolve only as a clean-room BLD memo benchmark. Extract memo-practice behavior into a scrubbed spec; do not recommend direct source incorporation.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Reference only","Retain only as benchmark, oracle, acceptance-test, or clean-room planning input.","exclude from donor lane","The repo has no explicit license visible and does not justify donor treatment; retain only as a clean-room memo benchmark.","medium","The clean-room benchmark role is clear, but the repo is narrow and license posture remains non-permissive in practice.","medium","The clean-room benchmark role is clear, but the repo is narrow and license posture remains non-permissive in practice.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","ambisinister/blindsolve — no explicit license visible; keep only as a clean-room BLD memo benchmark.","ambisinister/blindsolve is retained only as a clean-room BLD memo benchmark. Its value is in a narrow behavior slice, not in donor code.","memo","False","True","0.0","1.0","HT_training_clean_room","HT_training_clean_room_0002","Clean-room BLD memo benchmark","Clean-room BLD memo benchmark for HyperTwist; retained for behavior extraction only.","HyperTwist","Clean-room BLD memo benchmark","future candidate","architecture only","medium","ambisinister/blindsolve","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","no explicit license visible","pending_repo_license_audit","not_resolved_from_uploaded_materials","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","28.0","5","117.0","Reserve Bench","Clean-room BLD memo benchmark","Included","P3","license_unknown_pending","reverse_engineer_preferred","No explicit permissive license is visible. Retain this repo only as clean-room BLD memo benchmark material and implement any strategically necessary behavior from scrubbed first-party specifications.","Model A may inspect and summarize memo-practice behavior; Model B should implement only from a scrubbed clean-room specification.","Do not incorporate source directly without a confirmed license grant.","Yes — this is the preferred path if a memo-practice micro-mode later proves strategically necessary.","medium","no-license-clean-room-benchmark","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed clean-room BLD memo benchmark status.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as a no-license BLD memo benchmark for hide/reveal, letter-pair rehearsal, and feedback cadence; keep it narrow and clean-room only." "11930","efrantar/rob-twophase","https://github.com/efrantar/rob-twophase","HyperTwist","29.0","117.0","132.0","B","solver oracle / robot-metric benchmark","benchmark","Solver oracle and robot-metric benchmark","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Benchmark-only retained value: correctness oracle, robot/axial-metric comparisons, multiple-solution search expectations, and pruning-table performance baselines.","Keep as a restrictive solver oracle benchmark. Use it for correctness, metric comparison, and performance expectations; do not treat it as donor code.","Do not merge this repo into the HyperTwist core. Translate only oracle behavior and robot-metric expectations into first-party acceptance tests.","Repurpose here means: use as a correctness/performance oracle and robot-metric benchmark.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","oracle benchmark review","Hidden value sits in metric handling, pruning-table organization, CLI oracle behavior, and multi-solution search expectations.","Which oracle surfaces, robot metrics, and correctness/performance baselines should be preserved for first-party solver validation?","Inspect metric handling, pruning-table structure, CLI oracle behavior, and multiple-solution search as benchmark material only.","Audit efrantar/rob-twophase only as a restrictive solver oracle and robot-metric benchmark. Extract acceptance-test expectations and oracle behavior, not donor code.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Reference only","Retain only as benchmark, oracle, acceptance-test, or clean-room planning input.","exclude from donor lane","The repo is GPL solver code and should remain only as a correctness/performance oracle and robot-metric benchmark.","high","Dossier-backed classification is stable: the retained value is oracle and metric benchmarking, not direct incorporation.","high","Dossier-backed classification is stable: the retained value is oracle and metric benchmarking, not direct incorporation.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","efrantar/rob-twophase — GPL two-phase solver; keep only as solver oracle and robot-metric benchmark.","efrantar/rob-twophase remains valuable because it is a strong correctness/performance oracle with robot-metric coverage. That value is benchmark authority, not donor code.","memo","False","True","0.0","1.0","HT_solver_oracles","HT_solver_oracles_0002","Solver oracle and robot-metric benchmark","Solver oracle and robot-metric benchmark for HyperTwist; retained for acceptance testing, not donor use.","HyperTwist","Solver oracle and robot-metric benchmark","future candidate","architecture only","medium","efrantar/rob-twophase","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","GPL-3.0","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","29.0","5","117.0","Reserve Bench","Solver oracle and robot-metric benchmark","Included","P2","mixed_or_boundary_sensitive_known","pattern_only_preferred","The repo is GPL solver/oracle code and remains a secondary metric, pruning, and performance benchmark rather than donor code.","Reference only: use as a solver oracle and robot-metric benchmark without direct source incorporation.","Direct incorporation would require GPL-compatible distribution/compliance and is not the planned HyperTwist path.","Only if a narrow oracle behavior later needs first-party recreation; otherwise keep this as a benchmark.","high","gpl-solver-oracle-benchmark","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed solver oracle benchmark status.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as a secondary solver oracle for metric, pruning, and performance comparison; do not treat as donor code." "11933","onionhoney/roux-trainers","https://github.com/onionhoney/roux-trainers","HyperTwist","30.0","117.0","132.0","B","method-specific training donor","restrictive donor","clean-room donor","donor bench","repurpose","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Focused restrictive clean-room donor target. Retain the stage-specific training modes, solver-backed analyzer behavior, recognition masking, favorites/batch-drill flow, and 2D/3D training visualization ideas through Model A / Model B separation.","Keep in restrictive custody and route value through clean-room extraction only. Use it as a method-specific training and analyzer donor, not as direct donor code.","Do not merge this repo into the HyperTwist core as source. Convert the valuable training-state-machine and analyzer behaviors into first-party subsystems behind a clean-room boundary.","Repurpose here means: translate stage-specific training and analyzer behaviors into first-party implementations through a clean-room process.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","deep source audit","Hidden value sits in state-machine-driven training flows, solver-backed analyzer behavior, recognition masking, favorites persistence, and batch-drill mechanics.","Which Roux-specific training and analyzer behaviors are strategically worth reproducing in first-party code through a clean-room handoff?","Inspect blockbuilding micro-trainers, stage-specific state machines, solver-backed analyzer logic, recognition masking, favorites/batch-drill persistence, and 2D/3D visualization boundaries.","Audit onionhoney/roux-trainers as a focused restrictive clean-room donor for HyperTwist. Preserve method-specific training-state-machine, analyzer, recognition-masking, and drill-loop behavior in a scrubbed Model A handoff only.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","HyperTwist only","Restrictive clean-room donor value is real, but it should flow through Model A to a first-party implementation rather than through direct source sharing.","exclude from direct code use, keep as clean-room donor","The repo is GPLv3 and too method-specific for direct incorporation, but it still contains real training/analyzer subsystem value worth preserving through a clean-room donor path.","high","The dossier-backed classification is stable: stronger than a passive benchmark, but clearly a restrictive clean-room donor rather than direct donor code.","high","The dossier-backed classification is stable: stronger than a passive benchmark, but clearly a restrictive clean-room donor rather than direct donor code.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","onionhoney/roux-trainers — GPLv3 Roux trainer; keep as a focused restrictive clean-room donor, not direct donor code.","onionhoney/roux-trainers remains valuable because it contains stage-specific training-state machines, solver-backed analyzer modes, recognition masking, and batch-drill patterns. That value should flow through a clean-room donor path, not direct incorporation.","memo","False","True","0.0","1.0","HT_training_clean_room","HT_training_clean_room_0006","Focused restrictive clean-room donor target","Focused restrictive clean-room donor target for HyperTwist; stronger than a passive benchmark, but still not direct donor code.","HyperTwist","Focused restrictive clean-room donor target","repurpose","architecture only","medium","onionhoney/roux-trainers","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","GPL-3.0","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","30.0","5","117.0","Donor Bench","Focused restrictive clean-room donor target","Included","P2","mixed_or_boundary_sensitive_known","reverse_engineer_preferred","The repo is GPLv3 and should not be used as direct donor code in the HyperTwist core. Its value is in selective behavior and subsystem extraction through a clean-room Model A / Model B process.","Model A inspects the restrictive source; Model B implements only from a scrubbed first-party specification and must not access the repo directly.","Direct incorporation would require GPL-compatible distribution/compliance and is not the planned HyperTwist path.","Yes — preferred path for selectively reproducing stage-specific training and analyzer behavior in first-party code.","high","gpl-clean-room-donor","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed focused restrictive clean-room donor status.","implemented_live_clean_room_verified","landed_clean_room_preserve","Phase 1R closed. Preserve as the landed restrictive clean-room precedent; keep the Model A/Model B chain explicit and work only from first-party outputs or scrubbed specs." -"11936","yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","31.0","117.0","132.0","B","vision / perception / AR","subsystem donor","vision donor","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving calibration/detection/state-reconstruction logic while replacing camera UX and integration surfaces.","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","Integrate as a cubing / algorithm training subsystem for HyperTwist. Preserve the strongest existing pieces — scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition — and expose them behind a portfolio-stable interface. Wire first into tao-yu/Alg-Trainer, then into poliva/cubedex for orchestration, visualization, or data exchange.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Benchmark calibration pipeline; Extract state reconstruction; Wrap with camera/AR adapter","deep source audit","Hidden value often sits in calibration, preprocessing, stabilization, object/state reconstruction, replay artifacts, and camera-to-domain state pipelines that are not obvious from demos.","Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit yakupbilen/drl-rubiks-cube as a vision / perception / AR candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / vision donor. Test the three merger paths in order: 1) kkoomen/qbr [foundation + perception donor]; 2) vivaansinghvi07/rubix-cube-solver [perception + replay donor]; 3) cubing/cubing.js [state/render backend]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","kkoomen/qbr","foundation + perception donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","vivaansinghvi07/rubix-cube-solver","perception + replay donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","cubing/cubing.js","state/render backend","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from standalone? no","Not necessarily the final base, but keep as an active subsystem candidate or major donor.","medium","single-source signal; clear taxonomy; active integration value","medium","memo mentions: 1","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","yakupbilen/drl-rubiks-cube is treated as a future candidate for HyperTwist because the current dossier retains it mainly for learned-heuristic search, batched A-star style experimentation, and training-loop ideas, not as part of the committed perception core.","memo","False","True","0.0","1.0","HT_cube_vision","HT_cube_vision_0008","cubing / algorithm training","Integrate primarily for HyperTwist. Its memo and bookmark signals place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","yakupbilen/drl-rubiks-cube","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","31.0","5","117.0","Reserve Bench","Search/training systems bench","","","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." +"11936","yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","31.0","117.0","132.0","B","learned heuristic search / offline solver experimentation","research benchmark","solver experimentation benchmark","research benchmark","future candidate","architecture only","Retain the valuable internal research engine, but keep it below live product ownership. For this repo class, preserve sticker-state transition encoding, ADI state generation, batched neural search, and offline experiment loops while excluding camera UX, PyQt shell behavior, and near-term product runtime assumptions.","Benchmark-only retained value: 54-sticker transition encoding, ADI training-state generation, batched neural A* search, and compact value-network experimentation worth preserving as research reference.","Keep only as a research benchmark. Use it to shape first-party learned heuristic search, ADI state-generation, and offline solver experimentation if HyperTwist ever opens that lane; do not treat it as a live camera, calibration, recognition, or product-shell donor.","Keep behind the landed qbr and rubix-cube-solver recognition owners and behind the retained brownan/efrantar solver-oracle owners. Revisit only if HyperTwist intentionally opens a first-party ML search or offline experimentation lane.","Repurpose here means: mine cube-state transition encoding, ADI target-generation loops, batched heuristic-search experiments, and small value-network research rather than camera UX or product shell behavior.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Benchmark calibration pipeline; Extract state reconstruction; Wrap with camera/AR adapter","deep source audit","Hidden value often sits in calibration, preprocessing, stabilization, object/state reconstruction, replay artifacts, and camera-to-domain state pipelines that are not obvious from demos.","Inspect cube-state encoding; move-transition tables; ADI state generation; batched A* evaluation; value-network shape; offline training loop; and experiment boundaries. Explicitly exclude camera/perception shell claims from owner judgment.","Inspect README/docs, cube-state model, move-transition tables, search loop, ADI generation, network architecture, training scripts, and experiment configuration. Look for reusable search research primitives rather than camera UX, PyQt shell flows, or live perception assumptions.","Audit yakupbilen/drl-rubiks-cube as a learned-search and offline experimentation benchmark for HyperTwist. Do not stop at README-level claims. Inspect cube-state encoding, move-transition tables, ADI generation, batched neural A* search, value-network shape, and experiment loops. Decide whether the repo remains a non-live research benchmark and identify only those primitives that would matter if a future first-party ML search lane is intentionally opened. Do not treat camera UX or PyQt shell behavior as owner evidence.","brownan/Rubiks-Cube-Solver","correctness oracle","Use brownan above this row for compact correctness and heuristic-table ownership; keep yakupbilen only as adjacent learned-search research.","efrantar/rob-twophase","performance comparator","Use efrantar above this row for metric, pruning, and performance comparison; keep yakupbilen only as offline learned-search experimentation.","kkoomen/qbr","live recognition boundary check","Use qbr as the landed recognition/calibration owner and keep yakupbilen explicitly out of live perception-shell ownership.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from standalone? no","Not necessarily the final base, but keep as an active subsystem candidate or major donor.","medium","single-source signal; retained research value; not a live owner lane","medium","memo mentions: 1","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","yakupbilen/drl-rubiks-cube is treated as a future candidate for HyperTwist because the current dossier retains it mainly for learned-heuristic search, batched A-star style experimentation, and training-loop ideas, not as part of the committed perception core.","memo","False","True","0.0","1.0","HT_cube_vision","HT_cube_vision_0008","solver research / experimentation","Retain only as an MIT research benchmark for learned heuristic search, ADI state generation, and offline solver experimentation. Its strongest current value is research logic, not live perception or product shell ownership.","HyperTwist","solver research / experimentation","integrate","moderate modification","medium","yakupbilen/drl-rubiks-cube","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","31.0","5","117.0","Reserve Bench","Search/training systems bench","","","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." "11939","Hypercubers/hypercubing.xyz","https://github.com/Hypercubers/hypercubing.xyz","HyperTwist","32.0","92.0","101.0","C","hypercubing / nD engine","subsystem donor","simulation donor","locked strategic donor","repurpose","moderate modification","Treat this as a mineable codebase: keep selected internals (algorithms, renderers, adapters, parsers, schedulers) while replacing the surrounding product assumptions and architecture. For this repo class, that usually means preserving generalized puzzle/state/render logic while building a new application shell around it.","Website for the Hypercubing community","Repurpose selected subsystems rather than the whole product. Mine the repo for nD state model, move notation, renderer, projection controls, solver/traversal logic, puzzle serialization, replay; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Repurpose here means: turn it into a higher-dimensional renderer/simulator donor and shared interaction grammar for HyperTwist and long-horizon VectorShell.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Isolate puzzle/state core; Extract render/input abstractions; Document notation and save format","deep source audit","Hidden value is likely in generalized puzzle/state representations, higher-dimensional transforms, notation systems, save formats, puzzle generators, and rendering abstractions.","Inspect generalized puzzle/state model; notation parser; transform math; rendering abstraction; save/load format; puzzle generator; input mapping; performance optimizations.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for higher-dimensional state/notation representations, projection math, renderer abstractions, puzzle serialization, controls, and replay/training hooks.","Audit Hypercubers/hypercubing.xyz as a hypercubing / nD engine candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect generalized puzzle/state model; notation parser; transform math; rendering abstraction; save/load format; puzzle generator; input mapping; performance optimizations. Decide whether the best extraction path is heavy modification and whether it belongs as subsystem donor / simulation donor. Test the three merger paths in order: 1) HactarCE/Hyperspeedcube [foundation + donor]; 2) cubing/cubing.js [3D engine + notation/state donor]; 3) tao-yu/Alg-Trainer [training UX donor]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","HactarCE/Hyperspeedcube","foundation + donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","cubing/cubing.js","3D engine + notation/state donor","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","tao-yu/Alg-Trainer","training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium-low","single-source signal; clear taxonomy; mostly donor/reference role","medium","explicit bookmark description available","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","HyperTwist(1)","HyperTwist Consider","Website for the Hypercubing community","","Hypercubers/hypercubing.xyz is treated as repurpose for HyperTwist because visible metadata points to the hypercubing / nD simulation layer. Surface signal: Website for the Hypercubing community The fit looks real, but more as a donor/augmenter than as a standalone foundation.","bookmarks","True","False","1.0","0.0","HT_hyper_engine","HT_hyper_engine_0006","hypercubing / nD simulation","Integrate primarily for HyperTwist. Its description and bookmark placement under ""HyperTwist Consider"" place it in the hypercubing / nD simulation layer.","HyperTwist","hypercubing / nD simulation","integrate","heavy modification","medium","hypercubers/hypercubing.xyz","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","32.0","5","92.0","Locked Strategic Donor","Knowledge and curriculum donor","","P1","permissive_or_noncopyleft_known","direct_incorporation_ok","The repo is MIT and is retained as a high-value knowledge and curriculum donor. Direct use of code/content structures is legally straightforward where it materially helps HyperTwist.","Use selectively as a donor for knowledge structures, notation, taxonomy, and leaderboard-generation logic; do not confuse the site snapshot with the canonical repo.","Preserve MIT notices and attribution where required.","Usually unnecessary unless later replacing a narrow implementation seam is cleaner than carrying the upstream code.","high","permissive-knowledge-donor","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale unknown-license donor-bench posture to dossier-backed MIT knowledge/curriculum donor status.","implemented_live_permissive","landed_permissive_preserve","Phase 3R-B closed. Preserve as the landed first-party hypercubing knowledge, notation, progression, software-reference, and leaderboard-contract lane; start future widening from the live-lane audit, then the 3R-B implementation packet, then REPO_LICENSE_TRACKING.md." "11945","Aarav2709/KubeTimr","https://github.com/Aarav2709/KubeTimr","HyperTwist","34.0","88.0","97.0","C","cubing trainer / solver / timing","subsystem donor","training donor","donor bench","repurpose","moderate modification","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Aarav2709/KubeTimr — MIT — Keyboard-first offline timer. | Aarav2709/KubeTimr — MIT — Offline timer.","Repurpose selected subsystems rather than the whole product. Mine the repo for scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","deep source audit","Hidden value often sits in cube-state representation, scramble generation, weighted drill scheduling, recognition datasets, replay/timer internals, and case database schemas.","Inspect cube-state model; scramble generator; trainer weighting/scheduling; case database and metadata; replay/timer model; import/export of alg sets; smartcube or sensor adapters.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit Aarav2709/KubeTimr as a cubing trainer / solver / timing candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect cube-state model; scramble generator; trainer weighting/scheduling; case database and metadata; replay/timer model; import/export of alg sets; smartcube or sensor adapters. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / training donor. Test the three merger paths in order: 1) tao-yu/Alg-Trainer [base + donor swarm]; 2) poliva/cubedex [specialized training UX donor]; 3) Lykos/cube_trainer [sampling/analytics donor]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","project-local first","Cross-project transfer is possible, but the value is clearest inside the assigned project until source audit exposes more reusable primitives.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium","single-source signal; clear taxonomy; mostly donor/reference role","medium","strong adjacency to identified core stack; memo mentions: 2","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","Aarav2709/KubeTimr — MIT — Keyboard-first offline timer. | Aarav2709/KubeTimr — MIT — Offline timer.","Aarav2709/KubeTimr is treated as repurpose for HyperTwist because the current dossier keeps it as a focused subsystem donor for timer-state logic, split-phase handling, local persistence, rolling stats, and keyboard-first offline practice flow rather than as a broader training-platform anchor.","memo","False","True","0.0","2.0","HT_training_stack","HT_training_stack_0015","cubing / algorithm training","Integrate primarily for HyperTwist. Its memo and bookmark signals place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","aarav2709/kubetimr","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","34.0","5","88.0","Donor Bench","Focused subsystem donor","","","permissive_or_noncopyleft_known","direct_incorporation_ok","This repo is permissively licensed and currently best treated as a focused subsystem donor for timer-state logic, split-phase handling, local persistence, and keyboard-first practice flow. Selective incorporation is legally straightforward, but the product shell should still be reshaped to fit HyperTwist.","Direct embed, vendored module, package dependency, or tightly integrated adapter as the architecture requires.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","routine-review-only","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy adjacency wording and stale clean-room boundary posture to dossier-backed focused subsystem donor on 2026-04-25.","implemented_live_permissive","landed_permissive_preserve","Phase 3R-A closed. Preserve as the landed first-party timer, inspection, splits, stats, persistence, and timer-scoped replay lane; start future widening from the live-lane audit, then the 3R-A implementation packet, then REPO_LICENSE_TRACKING.md." "11951","roice3/MagicTile","https://github.com/roice3/MagicTile","HyperTwist","36.0","88.0","97.0","C","infra / runtime / observability / integration","subsystem donor","integration utility","locked strategic donor","repurpose","moderate modification","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving collectors/adapters/runtime topology logic while integrating into a larger control plane.","Non-euclidean Rubik's Cube Analogues","Repurpose selected subsystems rather than the whole product. Mine the repo for scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract collectors/adapters; Normalize topology/config schema; Bridge into control-plane API","deep source audit","Hidden value usually sits in service topology schemas, collector agents, auth/integration adapters, caching, deployment abstractions, and metrics/event correlation.","Inspect topology/service schema; collectors/agents; auth/integration adapters; caching/state sync; deployment/runtime abstractions; metrics/event correlation; config layering.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit roice3/MagicTile as a infra / runtime / observability / integration candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect topology/service schema; collectors/agents; auth/integration adapters; caching/state sync; deployment/runtime abstractions; metrics/event correlation; config layering. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / integration utility. Test the three merger paths in order: 1) project-local anchor [base + donor]; 2) shared portfolio utility [augmenter]; 3) cross-project transfer candidate [future merger]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","project-local anchor","base + donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","shared portfolio utility","augmenter","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","cross-project transfer candidate","future merger","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","project-local first","Cross-project transfer is possible, but the value is clearest inside the assigned project until source audit exposes more reusable primitives.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium-low","single-source signal; clear taxonomy; mostly donor/reference role","medium","explicit bookmark description available","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","HyperTwist(1)","HyperTwist Consider","Non-euclidean Rubik's Cube Analogues","","roice3/MagicTile is treated as repurpose for HyperTwist because visible metadata points to the cubing / algorithm training layer. Surface signal: Non-euclidean Rubik's Cube Analogues The fit looks real, but more as a donor/augmenter than as a standalone foundation.","bookmarks","True","False","1.0","0.0","HT_hyper_engine","HT_hyper_engine_0007","cubing / algorithm training","Integrate primarily for HyperTwist. Its description and bookmark placement under ""HyperTwist Consider"" place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","roice3/magictile","HyperTwist","","","","","","","","","","","","","","","","Original global operational v3 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","36.0","5","88.0","Locked Strategic Donor","Top-tier non-Euclidean geometry and topology donor","","P1","permissive_or_noncopyleft_known","direct_incorporation_ok","The repo is MIT and is explicitly retained as a top-tier geometry/topology donor. Direct donor use is legally straightforward if the architecture benefits from it.","Direct donor use or bounded adapter extraction are both acceptable; choose the seam that best preserves the topology and twist infrastructure.","Preserve MIT notices and attribution where required.","Usually unnecessary unless later replacing a narrow seam is cleaner than carrying the upstream code.","high","permissive-top-tier-donor","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Assigned from legacy HY_misc to HT_hyper_engine during cluster normalization on 2026-04-25.","implemented_live_permissive","landed_permissive_preserve","Later Phase 6R-D/T preserve sequence closed for the bounded retained slice. Preserve as the landed first-party tiling topology, geometry-family, and transform-aware macro remapping lane; start future widening from ROADMAP.md, FEATURE_REGISTRY.md, REPO_LICENSE_TRACKING.md, and the repo-specific landed Phase 6R packets, while broader host-shell ownership stays deferred." diff --git a/docs/repo_portfolio_unified_phase_g_v6_3.csv b/docs/repo_portfolio_unified_phase_g_v6_3.csv index 7f4a830..35698c8 100644 --- a/docs/repo_portfolio_unified_phase_g_v6_3.csv +++ b/docs/repo_portfolio_unified_phase_g_v6_3.csv @@ -34,7 +34,7 @@ "28.0","ambisinister/blindsolve","https://github.com/ambisinister/blindsolve","HyperTwist","28.0","117.0","132.0","B","clean-room BLD memo benchmark","benchmark","Clean-room BLD memo benchmark","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Benchmark-only retained value: narrow memo-practice behavior and BLD training flow worth preserving through a clean-room spec.","Keep only as a clean-room benchmark. Use it to shape a first-party memo-practice micro-mode; do not incorporate source.","Do not treat this repo as a merge candidate. Convert useful memo-practice ideas into a scrubbed Model A spec only.","Repurpose here means: derive first-party BLD memo-practice behavior through a clean-room process.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","clean-room benchmark review","Hidden value sits in narrow memo-practice flow and user-feedback behavior.","Which memo-practice behaviors are worth preserving through a scrubbed Model A handoff?","Inspect memo-practice flow, prompt cadence, scoring/feedback behavior, and narrow training-state transitions as clean-room benchmark material only.","Audit ambisinister/blindsolve only as a clean-room BLD memo benchmark. Extract memo-practice behavior into a scrubbed spec; do not recommend direct source incorporation.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Reference only","Retain only as benchmark, oracle, acceptance-test, or clean-room planning input.","exclude from donor lane","The repo has no explicit license visible and does not justify donor treatment; retain only as a clean-room memo benchmark.","medium","The clean-room benchmark role is clear, but the repo is narrow and license posture remains non-permissive in practice.","medium","The clean-room benchmark role is clear, but the repo is narrow and license posture remains non-permissive in practice.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","ambisinister/blindsolve — no explicit license visible; keep only as a clean-room BLD memo benchmark.","ambisinister/blindsolve is retained only as a clean-room BLD memo benchmark. Its value is in a narrow behavior slice, not in donor code.","memo","False","True","0.0","1.0","HT_training_clean_room","HT_training_clean_room_0002","Clean-room BLD memo benchmark","Clean-room BLD memo benchmark for HyperTwist; retained for behavior extraction only.","HyperTwist","Clean-room BLD memo benchmark","future candidate","architecture only","medium","ambisinister/blindsolve","1.0","Reserve Bench","5.0","No explicit license visible and extremely narrow scope; retain only as clean-room benchmark material.","Clean-room BLD memo benchmark","Included","P3","Keep in canon only as a clean-room BLD memo benchmark for later first-party micro-mode work.","2.0","3.0","3.0","11928.0","130.0","","ambisinister/blindsolve","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","no explicit license visible","pending_repo_license_audit","not_resolved_from_uploaded_materials","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","28.0","3","5.0","117.0","license_unknown_pending","reverse_engineer_preferred","No explicit permissive license is visible. Retain this repo only as clean-room BLD memo benchmark material and implement any strategically necessary behavior from scrubbed first-party specifications.","Model A may inspect and summarize memo-practice behavior; Model B should implement only from a scrubbed clean-room specification.","Do not incorporate source directly without a confirmed license grant.","Yes — this is the preferred path if a memo-practice micro-mode later proves strategically necessary.","medium","no-license-clean-room-benchmark","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed clean-room BLD memo benchmark status.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as a no-license BLD memo benchmark for hide/reveal, letter-pair rehearsal, and feedback cadence; keep it narrow and clean-room only." "29.0","efrantar/rob-twophase","https://github.com/efrantar/rob-twophase","HyperTwist","29.0","117.0","132.0","B","solver oracle / robot-metric benchmark","benchmark","Solver oracle and robot-metric benchmark","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Benchmark-only retained value: correctness oracle, robot/axial-metric comparisons, multiple-solution search expectations, and pruning-table performance baselines.","Keep as a restrictive solver oracle benchmark. Use it for correctness, metric comparison, and performance expectations; do not treat it as donor code.","Do not merge this repo into the HyperTwist core. Translate only oracle behavior and robot-metric expectations into first-party acceptance tests.","Repurpose here means: use as a correctness/performance oracle and robot-metric benchmark.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","oracle benchmark review","Hidden value sits in metric handling, pruning-table organization, CLI oracle behavior, and multi-solution search expectations.","Which oracle surfaces, robot metrics, and correctness/performance baselines should be preserved for first-party solver validation?","Inspect metric handling, pruning-table structure, CLI oracle behavior, and multiple-solution search as benchmark material only.","Audit efrantar/rob-twophase only as a restrictive solver oracle and robot-metric benchmark. Extract acceptance-test expectations and oracle behavior, not donor code.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Reference only","Retain only as benchmark, oracle, acceptance-test, or clean-room planning input.","exclude from donor lane","The repo is GPL solver code and should remain only as a correctness/performance oracle and robot-metric benchmark.","high","Dossier-backed classification is stable: the retained value is oracle and metric benchmarking, not direct incorporation.","high","Dossier-backed classification is stable: the retained value is oracle and metric benchmarking, not direct incorporation.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","efrantar/rob-twophase — GPL two-phase solver; keep only as solver oracle and robot-metric benchmark.","efrantar/rob-twophase remains valuable because it is a strong correctness/performance oracle with robot-metric coverage. That value is benchmark authority, not donor code.","memo","False","True","0.0","1.0","HT_solver_oracles","HT_solver_oracles_0002","Solver oracle and robot-metric benchmark","Solver oracle and robot-metric benchmark for HyperTwist; retained for acceptance testing, not donor use.","HyperTwist","Solver oracle and robot-metric benchmark","future candidate","architecture only","medium","efrantar/rob-twophase","1.0","Reserve Bench","5.0","GPL solver/oracle repo whose retained value is correctness, performance, and metric benchmarking rather than donor use.","Solver oracle and robot-metric benchmark","Included","P2","Keep in canon as a restrictive solver oracle benchmark with robot-metric value.","2.0","3.0","3.0","11931.0","131.0","","efrantar/rob-twophase","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","GPL-3.0","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","29.0","3","5.0","117.0","mixed_or_boundary_sensitive_known","pattern_only_preferred","The repo is GPL solver/oracle code and remains a secondary metric, pruning, and performance benchmark rather than donor code.","Reference only: use as a solver oracle and robot-metric benchmark without direct source incorporation.","Direct incorporation would require GPL-compatible distribution/compliance and is not the planned HyperTwist path.","Only if a narrow oracle behavior later needs first-party recreation; otherwise keep this as a benchmark.","high","gpl-solver-oracle-benchmark","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed solver oracle benchmark status.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as a secondary solver oracle for metric, pruning, and performance comparison; do not treat as donor code." "30.0","onionhoney/roux-trainers","https://github.com/onionhoney/roux-trainers","HyperTwist","30.0","117.0","132.0","B","method-specific training donor","restrictive donor","clean-room donor","donor bench","repurpose","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Focused restrictive clean-room donor target. Retain the stage-specific training modes, solver-backed analyzer behavior, recognition masking, favorites/batch-drill flow, and 2D/3D training visualization ideas through Model A / Model B separation.","Keep in restrictive custody and route value through clean-room extraction only. Use it as a method-specific training and analyzer donor, not as direct donor code.","Do not merge this repo into the HyperTwist core as source. Convert the valuable training-state-machine and analyzer behaviors into first-party subsystems behind a clean-room boundary.","Repurpose here means: translate stage-specific training and analyzer behaviors into first-party implementations through a clean-room process.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","deep source audit","Hidden value sits in state-machine-driven training flows, solver-backed analyzer behavior, recognition masking, favorites persistence, and batch-drill mechanics.","Which Roux-specific training and analyzer behaviors are strategically worth reproducing in first-party code through a clean-room handoff?","Inspect blockbuilding micro-trainers, stage-specific state machines, solver-backed analyzer logic, recognition masking, favorites/batch-drill persistence, and 2D/3D visualization boundaries.","Audit onionhoney/roux-trainers as a focused restrictive clean-room donor for HyperTwist. Preserve method-specific training-state-machine, analyzer, recognition-masking, and drill-loop behavior in a scrubbed Model A handoff only.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","HyperTwist only","Restrictive clean-room donor value is real, but it should flow through Model A to a first-party implementation rather than through direct source sharing.","exclude from direct code use, keep as clean-room donor","The repo is GPLv3 and too method-specific for direct incorporation, but it still contains real training/analyzer subsystem value worth preserving through a clean-room donor path.","high","The dossier-backed classification is stable: stronger than a passive benchmark, but clearly a restrictive clean-room donor rather than direct donor code.","high","The dossier-backed classification is stable: stronger than a passive benchmark, but clearly a restrictive clean-room donor rather than direct donor code.","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","onionhoney/roux-trainers — GPLv3 Roux trainer; keep as a focused restrictive clean-room donor, not direct donor code.","onionhoney/roux-trainers remains valuable because it contains stage-specific training-state machines, solver-backed analyzer modes, recognition masking, and batch-drill patterns. That value should flow through a clean-room donor path, not direct incorporation.","memo","False","True","0.0","1.0","HT_training_clean_room","HT_training_clean_room_0006","Focused restrictive clean-room donor target","Focused restrictive clean-room donor target for HyperTwist; stronger than a passive benchmark, but still not direct donor code.","HyperTwist","Focused restrictive clean-room donor target","repurpose","architecture only","medium","onionhoney/roux-trainers","1.0","Donor Bench","5.0","GPLv3 method-specific trainer with real subsystem value that should be preserved through clean-room extraction rather than donor use.","Focused restrictive clean-room donor target","Included","P2","Keep in canon as a focused restrictive clean-room donor for method-specific training and analyzer behavior.","2.0","3.0","3.0","11934.0","132.0","","onionhoney/roux-trainers","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","GPL-3.0","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","30.0","3","5.0","117.0","mixed_or_boundary_sensitive_known","reverse_engineer_preferred","The repo is GPLv3 and should not be used as direct donor code in the HyperTwist core. Its value is in selective behavior and subsystem extraction through a clean-room Model A / Model B process.","Model A inspects the restrictive source; Model B implements only from a scrubbed first-party specification and must not access the repo directly.","Direct incorporation would require GPL-compatible distribution/compliance and is not the planned HyperTwist path.","Yes — preferred path for selectively reproducing stage-specific training and analyzer behavior in first-party code.","high","gpl-clean-room-donor","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed focused restrictive clean-room donor status.","implemented_live_clean_room_verified","landed_clean_room_preserve","Phase 1R closed. Preserve as the landed restrictive clean-room precedent; keep the Model A/Model B chain explicit and work only from first-party outputs or scrubbed specs." -"31.0","yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","31.0","117.0","132.0","B","vision / perception / AR","subsystem donor","vision donor","Reserve Bench","future candidate","architecture only","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving calibration/detection/state-reconstruction logic while replacing camera UX and integration surfaces.","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","Integrate as a cubing / algorithm training subsystem for HyperTwist. Preserve the strongest existing pieces — scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition — and expose them behind a portfolio-stable interface. Wire first into tao-yu/Alg-Trainer, then into poliva/cubedex for orchestration, visualization, or data exchange.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Benchmark calibration pipeline; Extract state reconstruction; Wrap with camera/AR adapter","deep source audit","Hidden value often sits in calibration, preprocessing, stabilization, object/state reconstruction, replay artifacts, and camera-to-domain state pipelines that are not obvious from demos.","Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit yakupbilen/drl-rubiks-cube as a vision / perception / AR candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / vision donor. Test the three merger paths in order: 1) kkoomen/qbr [foundation + perception donor]; 2) vivaansinghvi07/rubix-cube-solver [perception + replay donor]; 3) cubing/cubing.js [state/render backend]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","kkoomen/qbr","foundation + perception donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","vivaansinghvi07/rubix-cube-solver","perception + replay donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","cubing/cubing.js","state/render backend","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from standalone? no","Not necessarily the final base, but keep as an active subsystem candidate or major donor.","medium","single-source signal; clear taxonomy; active integration value","medium","memo mentions: 1","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","yakupbilen/drl-rubiks-cube is treated as a future candidate for HyperTwist because the current dossier retains it mainly for learned-heuristic search, batched A-star style experimentation, and training-loop ideas, not as part of the committed perception core.","memo","False","True","0.0","1.0","HT_cube_vision","HT_cube_vision_0008","cubing / algorithm training","Integrate primarily for HyperTwist. Its memo and bookmark signals place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","yakupbilen/drl-rubiks-cube","1.0","Reserve Bench","5.0","Useful MIT search and training benchmark for learned-heuristic search, state generation, and experimentation loops, but explicitly below the committed perception and training core.","Search/training systems bench","Included","P2","yakupbilen/drl-rubiks-cube is placed in Reserve Bench for HyperTwist because it best serves the 'Search/training systems bench' role; recommended action is 'future candidate' with repurposing scope 'architecture only'. Confidence is medium because this remains a metadata-level judgment until source audit confirms hidden modules, plugin points, adapters, or architectural strengths.","2.0","3.0","3.0","11937.0","133.0","","yakupbilen/drl-rubiks-cube","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","31.0","3","5.0","117.0","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." +"31.0","yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","31.0","117.0","132.0","B","learned heuristic search / offline solver experimentation","research benchmark","solver experimentation benchmark","research benchmark","future candidate","architecture only","Retain the valuable internal research engine, but keep it below live product ownership. For this repo class, preserve sticker-state transition encoding, ADI state generation, batched neural search, and offline experiment loops while excluding camera UX, PyQt shell behavior, and near-term product runtime assumptions.","Benchmark-only retained value: 54-sticker transition encoding, ADI training-state generation, batched neural A* search, and compact value-network experimentation worth preserving as research reference.","Keep only as a research benchmark. Use it to shape first-party learned heuristic search, ADI state-generation, and offline solver experimentation if HyperTwist ever opens that lane; do not treat it as a live camera, calibration, recognition, or product-shell donor.","Keep behind the landed qbr and rubix-cube-solver recognition owners and behind the retained brownan/efrantar solver-oracle owners. Revisit only if HyperTwist intentionally opens a first-party ML search or offline experimentation lane.","Repurpose here means: mine cube-state transition encoding, ADI target-generation loops, batched heuristic-search experiments, and small value-network research rather than camera UX or product shell behavior.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Benchmark calibration pipeline; Extract state reconstruction; Wrap with camera/AR adapter","deep source audit","Hidden value often sits in calibration, preprocessing, stabilization, object/state reconstruction, replay artifacts, and camera-to-domain state pipelines that are not obvious from demos.","Inspect cube-state encoding; move-transition tables; ADI state generation; batched A* evaluation; value-network shape; offline training loop; and experiment boundaries. Explicitly exclude camera/perception shell claims from owner judgment.","Inspect README/docs, cube-state model, move-transition tables, search loop, ADI generation, network architecture, training scripts, and experiment configuration. Look for reusable search research primitives rather than camera UX, PyQt shell flows, or live perception assumptions.","Audit yakupbilen/drl-rubiks-cube as a learned-search and offline experimentation benchmark for HyperTwist. Do not stop at README-level claims. Inspect cube-state encoding, move-transition tables, ADI generation, batched neural A* search, value-network shape, and experiment loops. Decide whether the repo remains a non-live research benchmark and identify only those primitives that would matter if a future first-party ML search lane is intentionally opened. Do not treat camera UX or PyQt shell behavior as owner evidence.","brownan/Rubiks-Cube-Solver","correctness oracle","Use brownan above this row for compact correctness and heuristic-table ownership; keep yakupbilen only as adjacent learned-search research.","efrantar/rob-twophase","performance comparator","Use efrantar above this row for metric, pruning, and performance comparison; keep yakupbilen only as offline learned-search experimentation.","kkoomen/qbr","live recognition boundary check","Use qbr as the landed recognition/calibration owner and keep yakupbilen explicitly out of live perception-shell ownership.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from standalone? no","Not necessarily the final base, but keep as an active subsystem candidate or major donor.","medium","single-source signal; retained research value; not a live owner lane","medium","memo mentions: 1","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","yakupbilen/drl-rubiks-cube — MIT — RL solver with PyQt5 + webcam input.","yakupbilen/drl-rubiks-cube is treated as a future candidate for HyperTwist because the current dossier retains it mainly for learned-heuristic search, batched A-star style experimentation, and training-loop ideas, not as part of the committed perception core.","memo","False","True","0.0","1.0","HT_cube_vision","HT_cube_vision_0008","solver research / experimentation","Retain only as an MIT research benchmark for learned heuristic search, ADI state generation, and offline solver experimentation. Its strongest current value is research logic, not live perception or product shell ownership.","HyperTwist","solver research / experimentation","integrate","moderate modification","medium","yakupbilen/drl-rubiks-cube","1.0","Reserve Bench","5.0","Useful MIT search and training benchmark for learned-heuristic search, state generation, and experimentation loops, but explicitly below the committed perception and training core.","Search/training systems bench","Included","P2","yakupbilen/drl-rubiks-cube is placed in Reserve Bench for HyperTwist because it best serves the 'Search/training systems bench' role; recommended action is 'future candidate' with repurposing scope 'architecture only'. Confidence is medium because this remains a metadata-level judgment until source audit confirms hidden modules, plugin points, adapters, or architectural strengths.","2.0","3.0","3.0","11937.0","133.0","","yakupbilen/drl-rubiks-cube","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","31.0","3","5.0","117.0","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." "32.0","Hypercubers/hypercubing.xyz","https://github.com/Hypercubers/hypercubing.xyz","HyperTwist","32.0","92.0","101.0","C","hypercubing / nD engine","subsystem donor","simulation donor","locked strategic donor","repurpose","moderate modification","Treat this as a mineable codebase: keep selected internals (algorithms, renderers, adapters, parsers, schedulers) while replacing the surrounding product assumptions and architecture. For this repo class, that usually means preserving generalized puzzle/state/render logic while building a new application shell around it.","Website for the Hypercubing community","Repurpose selected subsystems rather than the whole product. Mine the repo for nD state model, move notation, renderer, projection controls, solver/traversal logic, puzzle serialization, replay; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Repurpose here means: turn it into a higher-dimensional renderer/simulator donor and shared interaction grammar for HyperTwist and long-horizon VectorShell.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Isolate puzzle/state core; Extract render/input abstractions; Document notation and save format","deep source audit","Hidden value is likely in generalized puzzle/state representations, higher-dimensional transforms, notation systems, save formats, puzzle generators, and rendering abstractions.","Inspect generalized puzzle/state model; notation parser; transform math; rendering abstraction; save/load format; puzzle generator; input mapping; performance optimizations.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for higher-dimensional state/notation representations, projection math, renderer abstractions, puzzle serialization, controls, and replay/training hooks.","Audit Hypercubers/hypercubing.xyz as a hypercubing / nD engine candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect generalized puzzle/state model; notation parser; transform math; rendering abstraction; save/load format; puzzle generator; input mapping; performance optimizations. Decide whether the best extraction path is heavy modification and whether it belongs as subsystem donor / simulation donor. Test the three merger paths in order: 1) HactarCE/Hyperspeedcube [foundation + donor]; 2) cubing/cubing.js [3D engine + notation/state donor]; 3) tao-yu/Alg-Trainer [training UX donor]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","HactarCE/Hyperspeedcube","foundation + donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","cubing/cubing.js","3D engine + notation/state donor","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","tao-yu/Alg-Trainer","training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium-low","single-source signal; clear taxonomy; mostly donor/reference role","medium","explicit bookmark description available","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","HyperTwist(1)","HyperTwist Consider","Website for the Hypercubing community","","Hypercubers/hypercubing.xyz is treated as repurpose for HyperTwist because visible metadata points to the hypercubing / nD simulation layer. Surface signal: Website for the Hypercubing community The fit looks real, but more as a donor/augmenter than as a standalone foundation.","bookmarks","True","False","1.0","0.0","HT_hyper_engine","HT_hyper_engine_0006","hypercubing / nD simulation","Integrate primarily for HyperTwist. Its description and bookmark placement under ""HyperTwist Consider"" place it in the hypercubing / nD simulation layer.","HyperTwist","hypercubing / nD simulation","integrate","heavy modification","medium","hypercubers/hypercubing.xyz","1.0","Locked Strategic Donor","4.0","MIT knowledge/curriculum donor with canonical notation, progression, taxonomy, and leaderboard-generation value.","Knowledge and curriculum donor","Included","P1","Keep as one of the strongest non-runtime donors in the hypercubing half of HyperTwist; the dossier-backed MIT posture and content value justify promotion above the old donor-bench treatment.","2.0","2.0","2.0","11938.0","134.0","Thin-fit assignment; verify project mapping during source audit","hypercubers/hypercubing.xyz","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","32.0","3","4.0","92.0","permissive_or_noncopyleft_known","direct_incorporation_ok","The repo is MIT and is retained as a high-value knowledge and curriculum donor. Direct use of code/content structures is legally straightforward where it materially helps HyperTwist.","Use selectively as a donor for knowledge structures, notation, taxonomy, and leaderboard-generation logic; do not confuse the site snapshot with the canonical repo.","Preserve MIT notices and attribution where required.","Usually unnecessary unless later replacing a narrow implementation seam is cleaner than carrying the upstream code.","high","permissive-knowledge-donor","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale unknown-license donor-bench posture to dossier-backed MIT knowledge/curriculum donor status.","implemented_live_permissive","landed_permissive_preserve","Phase 3R-B closed. Preserve as the landed first-party hypercubing knowledge, notation, progression, software-reference, and leaderboard-contract lane; start future widening from the live-lane audit, then the 3R-B implementation packet, then REPO_LICENSE_TRACKING.md." "34.0","Aarav2709/KubeTimr","https://github.com/Aarav2709/KubeTimr","HyperTwist","34.0","88.0","97.0","C","cubing trainer / solver / timing","subsystem donor","training donor","donor bench","repurpose","moderate modification","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","Aarav2709/KubeTimr — MIT — Keyboard-first offline timer. | Aarav2709/KubeTimr — MIT — Offline timer.","Repurpose selected subsystems rather than the whole product. Mine the repo for scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract cube-state model; Normalize trainer/case schema; Expose analytics and replay hooks","deep source audit","Hidden value often sits in cube-state representation, scramble generation, weighted drill scheduling, recognition datasets, replay/timer internals, and case database schemas.","Inspect cube-state model; scramble generator; trainer weighting/scheduling; case database and metadata; replay/timer model; import/export of alg sets; smartcube or sensor adapters.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit Aarav2709/KubeTimr as a cubing trainer / solver / timing candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect cube-state model; scramble generator; trainer weighting/scheduling; case database and metadata; replay/timer model; import/export of alg sets; smartcube or sensor adapters. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / training donor. Test the three merger paths in order: 1) tao-yu/Alg-Trainer [base + donor swarm]; 2) poliva/cubedex [specialized training UX donor]; 3) Lykos/cube_trainer [sampling/analytics donor]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","project-local first","Cross-project transfer is possible, but the value is clearest inside the assigned project until source audit exposes more reusable primitives.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium","single-source signal; clear taxonomy; mostly donor/reference role","medium","strong adjacency to identified core stack; memo mentions: 2","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","","","","Aarav2709/KubeTimr — MIT — Keyboard-first offline timer. | Aarav2709/KubeTimr — MIT — Offline timer.","Aarav2709/KubeTimr is treated as repurpose for HyperTwist because the current dossier keeps it as a focused subsystem donor for timer-state logic, split-phase handling, local persistence, rolling stats, and keyboard-first offline practice flow rather than as a broader training-platform anchor.","memo","False","True","0.0","2.0","HT_training_stack","HT_training_stack_0015","cubing / algorithm training","Integrate primarily for HyperTwist. Its memo and bookmark signals place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","aarav2709/kubetimr","1.0","Donor Bench","4.0","Useful subsystem donor for timer-state logic, split-phase handling, local persistence, rolling stats, and keyboard-first offline practice flow.","Focused subsystem donor","Included","P2","Aarav2709/KubeTimr is placed in Donor Bench for HyperTwist because it best serves the 'Focused subsystem donor' role; recommended action is 'repurpose' with repurposing scope 'moderate modification'. Confidence is medium because this remains a metadata-level judgment until source audit confirms hidden modules, plugin points, adapters, or architectural strengths.","2.0","3.0","2.0","11944.0","135.0","","aarav2709/kubetimr","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","34.0","3","4.0","88.0","permissive_or_noncopyleft_known","direct_incorporation_ok","This repo is permissively licensed and currently best treated as a focused subsystem donor for timer-state logic, split-phase handling, local persistence, and keyboard-first practice flow. Selective incorporation is legally straightforward, but the product shell should still be reshaped to fit HyperTwist.","Direct embed, vendored module, package dependency, or tightly integrated adapter as the architecture requires.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","routine-review-only","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy adjacency wording and stale clean-room boundary posture to dossier-backed focused subsystem donor on 2026-04-25.","implemented_live_permissive","landed_permissive_preserve","Phase 3R-A closed. Preserve as the landed first-party timer, inspection, splits, stats, persistence, and timer-scoped replay lane; start future widening from the live-lane audit, then the 3R-A implementation packet, then REPO_LICENSE_TRACKING.md." "36.0","roice3/MagicTile","https://github.com/roice3/MagicTile","HyperTwist","36.0","88.0","97.0","C","infra / runtime / observability / integration","subsystem donor","integration utility","locked strategic donor","repurpose","moderate modification","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving collectors/adapters/runtime topology logic while integrating into a larger control plane.","Non-euclidean Rubik's Cube Analogues","Repurpose selected subsystems rather than the whole product. Mine the repo for scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","Map module boundaries; Identify hidden reusable internals; Define adapter/API boundary to target anchors; Write extraction tests against upstream behavior; Extract collectors/adapters; Normalize topology/config schema; Bridge into control-plane API","deep source audit","Hidden value usually sits in service topology schemas, collector agents, auth/integration adapters, caching, deployment abstractions, and metrics/event correlation.","Inspect topology/service schema; collectors/agents; auth/integration adapters; caching/state sync; deployment/runtime abstractions; metrics/event correlation; config layering.","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Audit roice3/MagicTile as a infra / runtime / observability / integration candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect topology/service schema; collectors/agents; auth/integration adapters; caching/state sync; deployment/runtime abstractions; metrics/event correlation; config layering. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / integration utility. Test the three merger paths in order: 1) project-local anchor [base + donor]; 2) shared portfolio utility [augmenter]; 3) cross-project transfer candidate [future merger]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","project-local anchor","base + donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","shared portfolio utility","augmenter","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","cross-project transfer candidate","future merger","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","project-local first","Cross-project transfer is possible, but the value is clearest inside the assigned project until source audit exposes more reusable primitives.","exclude from core, keep as donor","Do not let it consume roadmap as a full product shell; mine reusable engines, adapters, schemas, UX patterns, or datasets.","medium-low","single-source signal; clear taxonomy; mostly donor/reference role","medium","explicit bookmark description available","Licensing intentionally ignored as a decision filter in this canonical evaluation; assess only architecture, capability, donor value, and concept transfer.","HyperTwist(1)","HyperTwist Consider","Non-euclidean Rubik's Cube Analogues","","roice3/MagicTile is treated as repurpose for HyperTwist because visible metadata points to the cubing / algorithm training layer. Surface signal: Non-euclidean Rubik's Cube Analogues The fit looks real, but more as a donor/augmenter than as a standalone foundation.","bookmarks","True","False","1.0","0.0","HT_hyper_engine","HT_hyper_engine_0007","cubing / algorithm training","Integrate primarily for HyperTwist. Its description and bookmark placement under ""HyperTwist Consider"" place it in the cubing / algorithm training layer.","HyperTwist","cubing / algorithm training","integrate","moderate modification","medium","roice3/magictile","1.0","Locked Strategic Donor","4.0","MIT donor with unusually strong non-Euclidean tiling, topology, and generalized twist infrastructure value.","Top-tier non-Euclidean geometry and topology donor","Included","P1","Keep as a top-tier non-Euclidean geometry and topology donor for HyperTwist; the dossier-backed MIT posture and source richness justify promotion above the old donor-bench treatment.","2.0","2.0","2.0","11950.0","136.0","Thin-fit assignment; verify project mapping during source audit","roice3/magictile","HyperTwist","","","","","","","","","","Original global Phase G v4 retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","36.0","3","4.0","88.0","permissive_or_noncopyleft_known","direct_incorporation_ok","The repo is MIT and is explicitly retained as a top-tier geometry/topology donor. Direct donor use is legally straightforward if the architecture benefits from it.","Direct donor use or bounded adapter extraction are both acceptable; choose the seam that best preserves the topology and twist infrastructure.","Preserve MIT notices and attribution where required.","Usually unnecessary unless later replacing a narrow seam is cleaner than carrying the upstream code.","high","permissive-top-tier-donor","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Assigned from legacy HY_misc to HT_hyper_engine during cluster normalization on 2026-04-25.","implemented_live_permissive","landed_permissive_preserve","Later Phase 6R-D/T preserve sequence closed for the bounded retained slice. Preserve as the landed first-party tiling topology, geometry-family, and transform-aware macro remapping lane; start future widening from ROADMAP.md, FEATURE_REGISTRY.md, REPO_LICENSE_TRACKING.md, and the repo-specific landed Phase 6R packets, while broader host-shell ownership stays deferred." diff --git a/docs/repo_portfolio_unified_source_audit_v6_3.csv b/docs/repo_portfolio_unified_source_audit_v6_3.csv index 296365a..a7045f9 100644 --- a/docs/repo_portfolio_unified_source_audit_v6_3.csv +++ b/docs/repo_portfolio_unified_source_audit_v6_3.csv @@ -193,7 +193,7 @@ "onionhoney/roux-trainers","https://github.com/onionhoney/roux-trainers","HyperTwist","Donor Bench","Focused restrictive clean-room donor target","P2","7481","9236","52.0","117.0","132.0","medium","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving cube-state, scramble, scheduling, or timer internals while adapting pedagogy, analytics, and UI.","repurpose","architecture only","Capture the stage-specific training, analyzer, recognition, and drill-loop behaviors that justify keeping roux-trainers as a clean-room donor target.","Stage-specific training flows, solver-backed analyzer logic, recognition masking, favorites/batch-drill persistence, and 2D/3D visualization boundaries.","Inspect blockbuilding micro-trainers, stage-specific state machines, solver-backed analyzer logic, recognition masking, favorites/batch-drill persistence, and 2D/3D visualization boundaries.","Which Roux-specific training and analyzer behaviors are strategically worth reproducing in first-party code through a clean-room handoff?","Keep in restrictive custody and route value through clean-room extraction only. Use it as a method-specific training and analyzer donor, not as direct donor code.","Do not merge this repo into the HyperTwist core as source. Convert the valuable training-state-machine and analyzer behaviors into first-party subsystems behind a clean-room boundary.","Repurpose here means: translate stage-specific training and analyzer behaviors into first-party implementations through a clean-room process.","tao-yu/Alg-Trainer","base + donor swarm","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","poliva/cubedex","specialized training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","Lykos/cube_trainer","sampling/analytics donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","HyperTwist only","Restrictive clean-room donor value is real, but it should flow through Model A to a first-party implementation rather than through direct source sharing.","Upgrade if source reveals clean modular architecture, reusable core abstractions, strong adapters/plugins, robust tests, and direct fit to a locked stack.","Demote if reusable value is mostly superficial, undocumented complexity overwhelms salvage value, or higher-ranked repos clearly dominate the same role.","Model A salvage map + clean-room boundary note + first-party subsystem extraction plan.","1) Which stage-specific subsystems are worth preserving 2) What must stay behind Model A / Model B separation 3) What first-party implementations should replace the GPL source","Primary: Phase G v4 board (canonical ranking and current bucket); Operational v3 board filtered to this repo; Merger matrix rows involving this repo; VS Code packet row for this repo, if present. Optional: Cluster narratives row for matching cluster; Bookmark occurrence rows for this repo. Do not attach v1/v2 or preliminary memo unless a judgment conflict, lineage ambiguity, or rationale gap needs arbitration.","Keep in canon as a focused restrictive clean-room donor for method-specific training and analyzer behavior.","GPLv3 method-specific trainer with real subsystem value that should be preserved through clean-room extraction rather than donor use.","Audit onionhoney/roux-trainers as a focused restrictive clean-room donor for HyperTwist. Preserve method-specific training-state-machine, analyzer, recognition-masking, and drill-loop behavior in a scrubbed Model A handoff only.","HT_training_clean_room_0006","HT_training_clean_room","onionhoney/roux-trainers","","","","","","","Original global P0-P3 source audit retained","GPL-3.0","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","2","3","117.0","mixed_or_boundary_sensitive_known","reverse_engineer_preferred","The repo is GPLv3 and should not be used as direct donor code in the HyperTwist core. Its value is in selective behavior and subsystem extraction through a clean-room Model A / Model B process.","Model A inspects the restrictive source; Model B implements only from a scrubbed first-party specification and must not access the repo directly.","Direct incorporation would require GPL-compatible distribution/compliance and is not the planned HyperTwist path.","Yes — preferred path for selectively reproducing stage-specific training and analyzer behavior in first-party code.","high","gpl-clean-room-donor","no","","","","","","","","","","","","","v6.3_final_source_of_truth","Corrected on 2026-04-25 from stale donor posture to dossier-backed focused restrictive clean-room donor status.","implemented_live_clean_room_verified","landed_clean_room_preserve","Phase 1R closed. Preserve as the landed restrictive clean-room precedent; keep the Model A/Model B chain explicit and work only from first-party outputs or scrubbed specs." -"yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","Reserve Bench","Adjacency / future transfer","P2","7482","9237","53.0","117.0","132.0","medium","Retain the valuable internal engine, but expect to replace UI/product shell, adapt schemas/APIs, and refactor boundaries so it can plug into the target anchors cleanly. For this repo class, that usually means preserving calibration/detection/state-reconstruction logic while replacing camera UX and integration surfaces.","future candidate","architecture only","Determine whether yakupbilen/drl-rubiks-cube should stay donor/merge-tier for HyperTwist, be promoted, or be demoted; identify concrete salvageable modules and best merge path.","puzzle model, learning flow, UX loops, data schema, replay/export, plugin/hooks","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for algorithm databases, spaced-repetition logic, scramble generation, timer/stat code, virtual cube components, smartcube adapters, and custom-trainer configuration support.","Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts.","Integrate as a cubing / algorithm training subsystem for HyperTwist. Preserve the strongest existing pieces — scramble generation, algorithm database, recognition/training loop, timing/statistics, virtual cube, smartcube hooks, spaced repetition — and expose them behind a portfolio-stable interface. Wire first into tao-yu/Alg-Trainer, then into poliva/cubedex for orchestration, visualization, or data exchange.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: tao-yu/Alg-Trainer, poliva/cubedex, Lykos/cube_trainer.","Repurpose here means: turn it into a trainer engine, solver/timer backend, recognition drill module, or method-specific practice mode for HyperTwist.","kkoomen/qbr","foundation + perception donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","vivaansinghvi07/rubix-cube-solver","perception + replay donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","cubing/cubing.js","state/render backend","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","Upgrade if source reveals clean modular architecture, reusable core abstractions, strong adapters/plugins, robust tests, and direct fit to a locked stack.","Demote if reusable value is mostly superficial, undocumented complexity overwhelms salvage value, or higher-ranked repos clearly dominate the same role.","Capability inventory + salvage targets + promotion/demotion verdict","1) Confirmed visible capabilities +"yakupbilen/drl-rubiks-cube","https://github.com/yakupbilen/drl-rubiks-cube","HyperTwist","Reserve Bench","Adjacency / future transfer","P2","7482","9237","53.0","117.0","132.0","medium","Retain the valuable internal research engine, but keep it below live product ownership. For this repo class, preserve sticker-state transition encoding, ADI state generation, batched neural search, and offline experiment loops while excluding camera UX, PyQt shell behavior, and near-term product runtime assumptions.","future candidate","architecture only","Determine whether yakupbilen/drl-rubiks-cube should stay donor/merge-tier for HyperTwist, be promoted, or be demoted; identify concrete salvageable modules and best merge path.","puzzle model, learning flow, UX loops, data schema, replay/export, plugin/hooks","Inspect README/docs, cube-state model, move-transition tables, search loop, ADI generation, network architecture, training scripts, and experiment configuration. Look for reusable search research primitives rather than camera UX, PyQt shell flows, or live perception assumptions.","Inspect cube-state encoding; move-transition tables; ADI state generation; batched A* evaluation; value-network shape; offline training loop; and experiment boundaries. Explicitly exclude camera/perception shell claims from owner judgment.","Keep only as a research benchmark. Use it to shape first-party learned heuristic search, ADI state-generation, and offline solver experimentation if HyperTwist ever opens that lane; do not treat it as a live camera, calibration, recognition, or product-shell donor.","Keep behind the landed qbr and rubix-cube-solver recognition owners and behind the retained brownan/efrantar solver-oracle owners. Revisit only if HyperTwist intentionally opens a first-party ML search or offline experimentation lane.","Repurpose here means: mine cube-state transition encoding, ADI target-generation loops, batched heuristic-search experiments, and small value-network research rather than camera UX or product shell behavior.","brownan/Rubiks-Cube-Solver","correctness oracle","Use brownan above this row for compact correctness and heuristic-table ownership; keep yakupbilen only as adjacent learned-search research.","efrantar/rob-twophase","performance comparator","Use efrantar above this row for metric, pruning, and performance comparison; keep yakupbilen only as offline learned-search experimentation.","kkoomen/qbr","live recognition boundary check","Use qbr as the landed recognition/calibration owner and keep yakupbilen explicitly out of live perception-shell ownership.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","Upgrade if source reveals clean modular architecture, reusable core abstractions, strong adapters/plugins, robust tests, and direct fit to a locked stack.","Demote if reusable value is mostly superficial, undocumented complexity overwhelms salvage value, or higher-ranked repos clearly dominate the same role.","Capability inventory + salvage targets + promotion/demotion verdict","1) Confirmed visible capabilities 2) Hidden capabilities found only in source 3) Best salvageable modules/files/packages 4) Integration path into target project @@ -201,7 +201,7 @@ 6) Best merge partners and exact coupling seam 7) Reasons to promote / retain / demote 8) Confidence change after source audit -9) Open questions / blockers","Primary: Phase G v4 board (canonical ranking and current bucket); Operational v3 board filtered to this repo; Merger matrix rows involving this repo; VS Code packet row for this repo, if present. Optional: Cluster narratives row for matching cluster; Bookmark occurrence rows for this repo. Do not attach v1/v2 or preliminary memo unless a judgment conflict, lineage ambiguity, or rationale gap needs arbitration.","yakupbilen/drl-rubiks-cube is placed in Reserve Bench for HyperTwist because it best serves the 'Search/training systems bench' role; recommended action is 'future candidate' with repurposing scope 'architecture only'. Confidence is medium because this remains a metadata-level judgment until source audit confirms hidden modules, plugin points, adapters, or architectural strengths.","Useful MIT search and training benchmark for learned-heuristic search, state generation, and experimentation loops, but explicitly below the committed perception and training core.","Audit yakupbilen/drl-rubiks-cube as a vision / perception / AR candidate for HyperTwist. Do not stop at README-level features. Inspect: Inspect preprocessing/calibration; tracking stabilization; state reconstruction; replay/event model; camera abstraction; fallback heuristics; testing assets/videos; performance shortcuts. Decide whether the best extraction path is moderate modification and whether it belongs as subsystem donor / vision donor. Test the three merger paths in order: 1) kkoomen/qbr [foundation + perception donor]; 2) vivaansinghvi07/rubix-cube-solver [perception + replay donor]; 3) cubing/cubing.js [state/render backend]. Return hidden modules, reusable schemas, protocol layers, plugin hooks, render/state models, datasets/test fixtures, and any subsystem stronger than the visible product shell.","HT_cube_vision_0008","HT_cube_vision","yakupbilen/drl-rubiks-cube","","","","","","","Original global P0-P3 source audit retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","2","3","117.0","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." +9) Open questions / blockers","Primary: Phase G v4 board (canonical ranking and current bucket); Operational v3 board filtered to this repo; Merger matrix rows involving this repo; VS Code packet row for this repo, if present. Optional: Cluster narratives row for matching cluster; Bookmark occurrence rows for this repo. Do not attach v1/v2 or preliminary memo unless a judgment conflict, lineage ambiguity, or rationale gap needs arbitration.","yakupbilen/drl-rubiks-cube is placed in Reserve Bench for HyperTwist because it best serves the 'Search/training systems bench' role; recommended action is 'future candidate' with repurposing scope 'architecture only'. Confidence is medium because this remains a metadata-level judgment until source audit confirms hidden modules, plugin points, adapters, or architectural strengths.","Useful MIT search and training benchmark for learned-heuristic search, state generation, and experimentation loops, but explicitly below the committed perception and training core.","Audit yakupbilen/drl-rubiks-cube as a learned-search and offline experimentation benchmark for HyperTwist. Do not stop at README-level claims. Inspect cube-state encoding, move-transition tables, ADI generation, batched neural A* search, value-network shape, and experiment loops. Decide whether the repo remains a non-live research benchmark and identify only those primitives that would matter if a future first-party ML search lane is intentionally opened. Do not treat camera UX or PyQt shell behavior as owner evidence.","HT_cube_vision_0008","HT_cube_vision","yakupbilen/drl-rubiks-cube","","","","","","","Original global P0-P3 source audit retained","MIT","known_from_reference_material","uploaded_reference_docs","no","","","","Supplemental intake references are advisory only; v6 adjudication remains the source of truth.","Usually indirect","yes","v6 unified all-project source-of-truth pack","v5_carry_forward","","v6_unified_source_of_truth_pack","2","3","117.0","permissive_or_noncopyleft_known","pattern_only_preferred","The repo is MIT and remains a research benchmark for learned heuristic search, ADI state generation, and offline experimentation rather than a near-term product donor.","Reference only: benchmark learned heuristic search and training experiments without near-term direct incorporation.","Typically preserve notices, attribution, and license text where required; no special copyleft-driven disclosure posture is normally needed.","Usually unnecessary unless you later decide the existing implementation is too constraining architecturally.","high","strategic-or-implemented-component","yes","","","","","","","","","","","","","v6.3_final_source_of_truth","Normalized legacy merge-bench wording to dossier-backed below-core benchmark posture on 2026-04-25.","not_live_reference_or_discard_candidate","phase0r_reference_benchmark_or_discard_eval","Packet 0R-E closed. Retain as an MIT research benchmark for learned heuristic search, ADI training loops, and offline experimentation rather than near-term product implementation." "Hypercubers/hypercubing.xyz","https://github.com/Hypercubers/hypercubing.xyz","HyperTwist","Locked Strategic Donor","Knowledge and curriculum donor","P1","7483","9238","1.0","92.0","101.0","medium","Treat this as a mineable codebase: keep selected internals (algorithms, renderers, adapters, parsers, schedulers) while replacing the surrounding product assumptions and architecture. For this repo class, that usually means preserving generalized puzzle/state/render logic while building a new application shell around it.","repurpose","moderate modification","Determine whether Hypercubers/hypercubing.xyz should stay donor/merge-tier for HyperTwist, be promoted, or be demoted; identify concrete salvageable modules and best merge path.","puzzle model, learning flow, UX loops, data schema, replay/export, plugin/hooks","Inspect package manifests, README/docs, src tree, examples, tests, CI workflows, config files, migrations/schemas, and hidden feature flags or experimental modules. Look for higher-dimensional state/notation representations, projection math, renderer abstractions, puzzle serialization, controls, and replay/training hooks.","Inspect generalized puzzle/state model; notation parser; transform math; rendering abstraction; save/load format; puzzle generator; input mapping; performance optimizations.","Repurpose selected subsystems rather than the whole product. Mine the repo for nD state model, move notation, renderer, projection controls, solver/traversal logic, puzzle serialization, replay; keep what materially shortens build time, but rebind data contracts, permissions, UI shell, storage, and deployment to the HyperTwist architecture. Best first pairing order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Consolidate under the Hyperspeedcube + cubing.js + qbr nucleus, not beside it as a separate silo. Normalize data contracts, auth/permissions, storage, and telemetry; then attach as a service, plugin, canvas layer, trainer, or analysis module. Descending combination order: HactarCE/Hyperspeedcube, cubing/cubing.js, tao-yu/Alg-Trainer.","Repurpose here means: turn it into a higher-dimensional renderer/simulator donor and shared interaction grammar for HyperTwist and long-horizon VectorShell.","HactarCE/Hyperspeedcube","foundation + donor","Use this repo against the partner as an augmenting layer; preserve the partner as the likely base and mine this repo for capabilities that improve breadth, UX, or specialization.","cubing/cubing.js","3D engine + notation/state donor","Use the partner for canonical state or rendering abstractions and merge this repo's specialized logic on top.","tao-yu/Alg-Trainer","training UX donor","Treat this pairing as a candidate donor merge rather than a replacement decision; inspect module boundaries to determine which side owns the final shell.","VectorShell | ScriptoriumAI","VectorShell can borrow spatial/rendering and perception primitives; ScriptoriumAI can borrow tutorial/educational visualization patterns rather than the full engine.","Upgrade if source reveals clean modular architecture, reusable core abstractions, strong adapters/plugins, robust tests, and direct fit to a locked stack.","Demote if reusable value is mostly superficial, undocumented complexity overwhelms salvage value, or higher-ranked repos clearly dominate the same role.","Capability inventory + salvage targets + promotion/demotion verdict","1) Confirmed visible capabilities 2) Hidden capabilities found only in source 3) Best salvageable modules/files/packages diff --git a/docs/v6_5_deep_manual_pack/HyperTwist/FEATURE_REGISTRY.md b/docs/v6_5_deep_manual_pack/HyperTwist/FEATURE_REGISTRY.md index 15888ec..a887658 100644 --- a/docs/v6_5_deep_manual_pack/HyperTwist/FEATURE_REGISTRY.md +++ b/docs/v6_5_deep_manual_pack/HyperTwist/FEATURE_REGISTRY.md @@ -190,6 +190,7 @@ repo. | Recognition-comparison adjunct boundary | Deep-source grounded retained | `cahidenes/rubiks-cube-solver` + `tentone/rubix-solver` retained adjunct lane | Narrow face-placement, cube-string cross-check, quad sorting, square-mask color sampling, center-color face labeling, and native face/state comparison value remain retained only behind the landed `qbr`, `rubix-cube-solver`, and first-party correction-stack owners. No default widening packet is open. | | Multi-face correction/explanation shell | Implemented now | first-party recognition/training surfaces | Current bounded correction profile, contradiction-aware correction state, accepted-resolution ledger, and correction-explanation / solve-unlock shell are live above the landed `Phase 6R-O`, `Phase 6R-P`, `Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S` seams. Bundled-asset, generalized solver-backend, and broad playback/runtime widening remain deferred. | | Compact classic-cube solver correctness and heuristic-search oracle | Deep-source grounded retained | `brownan/Rubiks-Cube-Solver` retained lane + `efrantar/rob-twophase` comparator | Broad solver-backend ownership remains non-live. `brownan` remains the primary compact correctness and heuristic-table oracle, while `efrantar` remains the retained metric, pruning, robot-execution, and performance comparator. | +| Learned heuristic search and `ADI` state-generation research benchmark | Deep-source grounded retained | `yakupbilen/drl-rubiks-cube` retained research lane | Retained only for `54`-sticker transition encoding, anti-cancellation move-sequence generation, `ADI` state generation, batched neural `A*` search experiments, and small value-network research. This does not displace `qbr` or `rubix-cube-solver` for live perception/reconstruction ownership, and it does not displace `brownan` or `efrantar` for correctness/performance oracle ownership. | ### 3. Training, coaching, and progression cockpit