Reconcile guided task planner hierarchy
This commit is contained in:
axiomlogicnexus
parent
b69413eaba
commit
46974024ad
4 changed files with 231 additions and 2 deletions
|
|
@ -0,0 +1,208 @@
|
|||
# HyperTwist Guided Classic-Cube Task-Planner And Focus-Playback Hierarchy Reconciliation - 2026-05-27
|
||||
|
||||
## Status
|
||||
|
||||
This document is the source-backed reconciliation note for the current guided
|
||||
classic-cube task-planner and focus-playback family across:
|
||||
|
||||
- `alinen/cube`
|
||||
- the first-party training, drill, coaching, and replay surfaces
|
||||
- the landed bounded classic-cube `Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S`
|
||||
seams
|
||||
|
||||
It answers one narrow question:
|
||||
|
||||
- does the current `alinen/cube` source require reopening already landed
|
||||
HyperTwist training/replay or classic-cube explanation packets before any
|
||||
future guided-task packet is considered
|
||||
|
||||
Result:
|
||||
|
||||
- no product-code reopening is required
|
||||
- yes selective authority backfill is required so the retained planner slice
|
||||
stays narrow and does not blur into the current first-party
|
||||
training/drill/coaching, replay, or bounded classic-cube explanation
|
||||
families
|
||||
|
||||
## 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\v6_5_deep_manual_pack\HyperTwist\FEATURE_REGISTRY.md`
|
||||
- `C:\Workspaces\HyperTwist\repos.manifest.json`
|
||||
- `C:\Workspaces\HyperTwist\clean-room-specs\alinen-cube.model-a.md`
|
||||
- `C:\Workspaces\HyperTwist\mirrors\restrictive\alinen\cube\README.md`
|
||||
- `C:\Workspaces\HyperTwist\mirrors\restrictive\alinen\cube\Assets\Scripts\CubePlanner.cs`
|
||||
- `C:\Workspaces\HyperTwist\mirrors\restrictive\alinen\cube\Assets\Scripts\CubeTaskSolver.cs`
|
||||
- `C:\Workspaces\HyperTwist\mirrors\restrictive\alinen\cube\Assets\Scripts\CubeStateManager.cs`
|
||||
- `C:\Workspaces\HyperTwist\mirrors\restrictive\alinen\cube\Assets\Scripts\CubeController.cs`
|
||||
- current first-party runtime and packet authority under:
|
||||
- `C:\HyperTwist\UnrealHyperTwist\Source\UnrealHyperTwist`
|
||||
- `C:\HyperTwist\docs\ops\HYPERTWIST_IMPLEMENTATION_PHASE_1_KICKOFF.md`
|
||||
|
||||
## Slice-local verdict
|
||||
|
||||
| Slice | Primary authority | Secondary value kept | Route | Width | Result |
|
||||
| --- | --- | --- | --- | --- | --- |
|
||||
| Landed first-party training sessions, method-drill definitions, run state, follow-up queues, and coaching cockpit surfaces | first-party training and coaching surfaces | none required from `alinen/cube` | existing landed permissive route | `F2` | unchanged live implementation owner |
|
||||
| Landed first-party replay and staged explanation shells | first-party replay surfaces plus landed bounded classic-cube `Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S` seams | none required from `alinen/cube` | existing landed permissive route | `F2` | unchanged live implementation owner |
|
||||
| Guided task graph, bounded task-local planner, focus-target cueing, and queued move demonstration for beginner classic-cube help | `alinen/cube` | current first-party training/coaching, replay, and bounded classic-cube explanation families remain adjacent non-overlapping owners | `R4` now, `R3` only if an explicit clean-room packet is opened later | `F2` | restrictive planner benchmark authority kept explicit |
|
||||
|
||||
## Why no product-code reopening is required
|
||||
|
||||
### `alinen/cube`
|
||||
|
||||
`alinen` still has retained value, but it is narrow:
|
||||
|
||||
- named solve-task progression rather than a single monolithic solve path
|
||||
- bounded per-task search and scoring against current constraints
|
||||
- explicit focus-target cueing for the current task
|
||||
- queued move demonstration over a visible cube surface
|
||||
- separable cube-state, planner, and playback modules
|
||||
|
||||
That retained value is explicit in the current source:
|
||||
|
||||
- `CubePlanner.cs` builds an ordered `_tasks` list with row/task labels and
|
||||
per-task solver delegates
|
||||
- `CubePlanner.cs` advances one task at a time and only requests bounded help
|
||||
against the current task
|
||||
- `CubePlanner.cs` uses `FocusCubie(...)` and task-specific color-enable logic
|
||||
to spotlight the current teaching target
|
||||
- `CubeTaskSolver.cs` scores candidate paths against the current cubie and
|
||||
solved constraints rather than only asking for a total-cube finish
|
||||
- `CubeController.cs` owns a queued command-playback seam separate from the
|
||||
planner
|
||||
- `CubeStateManager.cs` keeps the cube-state turn/grouping layer separate from
|
||||
both the planner and the playback shell
|
||||
|
||||
Those strengths make `alinen/cube` the strongest current guided task-planner
|
||||
benchmark. They do not make it the owner of HyperTwist's live training,
|
||||
coaching, replay, or bounded classic-cube explanation families.
|
||||
|
||||
### Current first-party training/coaching and replay surfaces
|
||||
|
||||
HyperTwist already owns the current live training, coaching, and adjacent
|
||||
replay family in first-party code:
|
||||
|
||||
- method-drill sample decks and definitions already exist
|
||||
- method-drill run, batch, diagnostic, and follow-up template surfaces already
|
||||
exist
|
||||
- coaching queue and follow-up entry surfaces already exist
|
||||
- replay and staged explanation surfaces already exist through first-party
|
||||
replay and bounded classic-cube explanation lanes
|
||||
|
||||
Those surfaces are broader and already live. `alinen/cube` does not reopen
|
||||
them.
|
||||
|
||||
### Landed classic-cube `Phase 6R-Q/R/S`
|
||||
|
||||
The landed classic-cube explanation and correction seams are already first-party
|
||||
and bounded:
|
||||
|
||||
- the stage ladder already exposes ordered solve stages
|
||||
- solve guidance already gates on reconstruction and correction completion
|
||||
- correction and resolution already reopen or unlock the stage ladder in a
|
||||
bounded way
|
||||
- playback-stage ownership is intentionally deferred inside that bounded family
|
||||
|
||||
`alinen/cube`'s task-planner benchmark does not reopen those shell/state
|
||||
families. It only remains a later clean-room comparison surface if HyperTwist
|
||||
ever wants a more explicit guided task graph above the current drill, replay,
|
||||
and solve-guidance seams.
|
||||
|
||||
## Selective backfill result
|
||||
|
||||
### Earlier packet reopened
|
||||
|
||||
- `C:\HyperTwist\docs\HYPERTWIST_PHASE_0R_PACKET_0R_E_EVALUATION_2026-05-13.md`
|
||||
|
||||
### Earlier standing note reopened
|
||||
|
||||
- `C:\Workspaces\HyperTwist\clean-room-specs\alinen-cube.model-a.md`
|
||||
|
||||
### Reason
|
||||
|
||||
Later comparison against the now-landed HyperTwist training/drill/coaching,
|
||||
replay, and classic-cube explanation surfaces sharpened the exact retained
|
||||
`alinen` slice:
|
||||
|
||||
- `alinen/cube` is not a generic training-session owner
|
||||
- `alinen/cube` is not a generic coaching backend owner
|
||||
- `alinen/cube` is not the live owner of canonical replay or staged
|
||||
explanation playback
|
||||
- `alinen/cube` is not the live owner of the current bounded solve-guidance
|
||||
ladder
|
||||
- `alinen/cube` remains the task-planner benchmark only for the narrower
|
||||
lesson-task graph, bounded task-local planner, focus-target cueing, and
|
||||
queued move-demonstration slice
|
||||
|
||||
### Change class
|
||||
|
||||
- repo-local extraction did not change
|
||||
- cross-lane authority wording was sharpened for the exact retained slice
|
||||
- no cleanup interpretation changed
|
||||
- no product-code change occurred
|
||||
|
||||
### What remained valid
|
||||
|
||||
- the original `0R-E` decision to keep `alinen/cube` no-license and non-live
|
||||
remained valid
|
||||
- the current first-party training/coaching surfaces remained valid
|
||||
- the current first-party replay surfaces remained valid
|
||||
- the current first-party classic-cube `Phase 6R-Q/R/S` seams remained valid
|
||||
- the existing scrubbed `Model A` handoff remained valid as a seed
|
||||
|
||||
### What changed
|
||||
|
||||
The dependent canon now makes the exact non-live role explicit:
|
||||
|
||||
- keep `alinen/cube` only as the strongest clean-room planner benchmark for
|
||||
the narrower guided task-graph and focus-playback slice
|
||||
- keep training/drill/coaching ownership with the current first-party surfaces
|
||||
- keep replay ownership with the current first-party replay family
|
||||
- keep bounded classic-cube explanation and correction ownership with the
|
||||
landed `Phase 6R-Q/R/S` seams
|
||||
|
||||
## Propagation result
|
||||
|
||||
### Feature registry
|
||||
|
||||
Updated.
|
||||
|
||||
The registry now explicitly records the retained guided task-planner and
|
||||
focus-playback slice instead of leaving that slice implicit in the older `0R-E`
|
||||
packet alone.
|
||||
|
||||
### Cross-lane doctrine
|
||||
|
||||
No doctrine wording change was needed.
|
||||
|
||||
This pass applied the existing `A/R/F` hierarchy; it did not change the
|
||||
hierarchy rules themselves.
|
||||
|
||||
### Skillization
|
||||
|
||||
No doctrine or `SKILLS.md` change was needed.
|
||||
|
||||
This pass did not create or reopen any command or skill surface.
|
||||
|
||||
### Legal evidence
|
||||
|
||||
No license-status or mirror-custody change occurred.
|
||||
|
||||
The repo-row tracker, feature registry, manifest, and scrubbed handoff were
|
||||
sharpened instead.
|
||||
|
||||
## Operational answer
|
||||
|
||||
Before any future guided task or planner packet is opened, the correct move was
|
||||
this evaluation and selective backfill pass first.
|
||||
|
||||
After this pass:
|
||||
|
||||
- do not reopen the live product code
|
||||
- do not reinterpret `alinen/cube` as proof that training/coaching, replay,
|
||||
or bounded solve-guidance ownership is still pending
|
||||
- if HyperTwist later opens a true guided task-planner lane, start from the
|
||||
retained planner benchmark slice only and keep it clean-room routed rather
|
||||
than donor-routed
|
||||
|
|
@ -595,6 +595,19 @@ Integration call:
|
|||
- retain as a clean-room planner and step-segmentation benchmark
|
||||
- if future teaching work needs planner-style explicit tasks, start from scrubbed task contracts rather than source reuse
|
||||
|
||||
2026-05-27 hierarchy overlay:
|
||||
|
||||
- `alinen/cube` remains `A1 + R4 + F2` only for the retained lesson-task
|
||||
graph, bounded task-local planner, focus-cue targeting, and queued
|
||||
move-demonstration slice
|
||||
- do not reinterpret the current first-party training-session, method-drill,
|
||||
coaching queue/follow-up, canonical replay, or bounded classic-cube
|
||||
`Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S` shell/state families through
|
||||
this row
|
||||
- use
|
||||
`C:\HyperTwist\docs\HYPERTWIST_GUIDED_CLASSIC_CUBE_TASK_PLANNER_AND_FOCUS_PLAYBACK_HIERARCHY_RECONCILIATION_2026-05-27.md`
|
||||
as the current comparative authority before later clean-room work
|
||||
|
||||
### `ambisinister/blindsolve`
|
||||
|
||||
Inspected source surfaces:
|
||||
|
|
|
|||
|
|
@ -1591,9 +1591,16 @@ Source basis:
|
|||
|
||||
Approved working posture:
|
||||
|
||||
- keep as a planner/task-segmentation clean-room benchmark
|
||||
- keep as an `A1 + R4 + F2` clean-room benchmark only for the retained
|
||||
lesson-task graph, bounded task-local planner, focus-target cueing, and
|
||||
queued move-demonstration slice
|
||||
- do not reinterpret the current first-party training-session, method-drill,
|
||||
coaching queue/follow-up, canonical replay, or bounded classic-cube
|
||||
`Phase 6R-Q`, `Phase 6R-R`, and `Phase 6R-S` shell/state families through
|
||||
this row
|
||||
- do not route direct donor work from this row
|
||||
- preserve the existing model-a file only as bounded seed material beneath the packet authority
|
||||
- preserve the existing model-a file only as bounded seed material beneath the
|
||||
packet authority
|
||||
|
||||
### `ambisinister/blindsolve`
|
||||
|
||||
|
|
|
|||
|
|
@ -182,6 +182,7 @@ repo.
|
|||
| Training sessions, method drills, and coaching dashboard | Implemented now | first-party training/coaching surfaces | Current live cockpit anchor. |
|
||||
| Follow-up templates, session queues, and coach summaries | Implemented now | first-party current code | Real first-party orchestration surface. |
|
||||
| Guided beginner lesson-state and stepwise pedagogy benchmark | Deep-source grounded retained | `AviKaufman/Rubix-cube-trainer` retained clean-room benchmark lane | Retained as `A1 + R4 + F2` only for named beginner milestones, oversatisfied-step explanation, bounded step help, and manual-turn/tutorial coherence on a live cube surface. This does not displace the first-party training-session, method-drill, coaching queue/follow-up, or bounded classic-cube `Phase 6R-Q/R/S` explanation and correction seams. |
|
||||
| Guided task-planner and focus-playback benchmark | Deep-source grounded retained | `alinen/cube` retained clean-room benchmark lane | Retained as `A1 + R4 + F2` only for a separable lesson-task graph, bounded task-local planner, focus-target cueing, and queued move demonstration on a visible cube surface. This does not displace the first-party training-session, method-drill, coaching queue/follow-up, canonical replay, or bounded classic-cube `Phase 6R-Q/R/S` explanation and correction seams. |
|
||||
| Embodied companion/narration adjunct | Implemented now | landed `TalkingHead` packet | Live bounded avatar/narration family. |
|
||||
| Broader long-horizon coach memory and cognitive state | Deep-source grounded retained | first-party target + memory doctrine | Doctrine-governed future widening; not yet a mature memory system. |
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue