EducationOS Pack Runner — 12-Week Schedule + Logging Template (Runnable) v1.0

(Paste-ready. Designed for tutor + student + parent. Low friction, high stability.)


Canonical One-Liner

Run a small set of corridors on a fixed cadence, log 5 sensors, repair the single highest-leak point each week, and only scale difficulty after transfer + variance gates pass.


0) Runner Metadata

RunnerID: EDU×Z0×Runner×Starter12Weeks×v1.0
Pack: EDU×Z0×Pack×Starter12×v1.0
Session length: 30–60 min/day + 1 longer weekly session
Non-negotiables: scope lock, plan-first, timed output, checklist sweep, weekly transfer variants, weekly calibration


1) Weekly Cadence (Never Change)

Each week has the same backbone to reduce choice overload:

Daily (Mon–Fri)

  • MicroOutput6 (10 min)
  • NodeBindExpand OR InputHarvest15 rotation (15–20 min)

Twice per week

  • ScopeLockQ + 5BoxPlan + TimedCoreOutput + ChecklistSweep (30–45 min)

Once per week

  • RewriteWeakestPara (20–30 min)
  • TransferVariant3 (20–30 min)
  • BacktestCalibrate (15–20 min)

Once per month (Weeks 4, 8, 12)

  • FullSimPaper (exam conditions)

2) 12-Week Schedule (Phase by Phase)

Weeks 1–3 — Phase 1 (Depth + Bind Foundation)

Goal: build usable nodes/binds + stabilise low-choice production.

Mon/Wed/Fri

  • MicroOutput6
  • NodeBindExpand (5 nodes)

Tue/Thu

  • MicroOutput6
  • InputHarvest15 (weekly target 15)

Sat (or longest session)

  • ScopeLockQ → 5BoxPlan → TimedCoreOutput → ChecklistSweep
  • RewriteWeakestPara (based on output)
  • BacktestCalibrate

Sun (short)

  • TransferVariant3 (light variants; focus on stability)

Weeks 4–6 — Phase 2 (Structure + Planning Discipline)

Goal: plan-first becomes automatic; scope drift drops.

Mon/Wed

  • MicroOutput6
  • NodeBindExpand (connectors + constraints)

Tue/Thu

  • MicroOutput6
  • ScopeLockQ + 5BoxPlan (fast drill, 10–12 min)
  • (Optional) ExplainBack3 (2 min)

Fri

  • InputHarvest15 (finish 15/week) + MicroOutput6

Sat

  • Full routine: ScopeLockQ → 5BoxPlan → TimedCoreOutput → ChecklistSweep
  • RewriteWeakestPara
  • BacktestCalibrate

Sun

  • TransferVariant3 (3 variants, scored)

Week 4 includes: FullSimPaper (replace one Sat run)


Weeks 7–9 — Phase 3 (Load Training + Variance Reduction)

Goal: output stabilises under time; variance drops; repair latency shortens.

Mon

  • MicroOutput6
  • ChecklistSweep drill (on yesterday’s paragraph)

Tue/Thu

  • ScopeLockQ → 5BoxPlan → TimedCoreOutput → ChecklistSweep

Wed/Fri

  • InputHarvest15 / NodeBindExpand rotation + MicroOutput6

Sat

  • RewriteWeakestPara (deep repair)
  • TransferVariant3
  • BacktestCalibrate

Week 8 includes: FullSimPaper


Weeks 10–12 — Phase 4 (Transfer + Exam Application)

Goal: transfer pass rate high; timed stability; exam simulation confidence.

Mon/Wed

  • MicroOutput6
  • TransferVariant3 (mini—1 variant each day)

Tue/Thu

  • Full timed run: ScopeLockQ → 5BoxPlan → TimedCoreOutput → ChecklistSweep

Fri

  • InputHarvest15 + NodeBindExpand (light)
  • BacktestCalibrate (short)

Sat

  • Full TransferVariant3 (all 3 variants)
  • RewriteWeakestPara (only weakest)
  • Log + calibrate

Week 12 includes: FullSimPaper (final benchmark)


3) The 5-Sensor Logging Template (Minimal, Mandatory)

Log after every “timed run” and every transfer session.

Log fields

  • Time: minutes used / allowed
  • Variance marker: stable or unstable? (1–5)
  • Transfer: pass/fail (and why)
  • Top failure mode: scope / bind gap / branch explosion / timing / retrieval
  • Repair latency: minutes to fix weakest part (or “not fixed”)

Optional (nice-to-have)

  • ForkCount (how many times you changed direction mid-run)
  • ConnectorAccuracy (0–3)

4) Weekly Calibration Rule (One Change Only)

During BacktestCalibrate each week:

  1. Pick the #1 failure mode (most frequent)
  2. Pick the #1 breached sensor (variance/transfer/time/repair latency)
  3. Change one thing only next week:
  • add 1 extra RewriteWeakestPara
  • add 1 extra TransferVariant
  • reduce prompt difficulty
  • add buffer time
  • enforce strict plan adherence

No more than one change per week (prevents Visionary Storm inside learning).


5) Gates (When to Scale)

You may scale difficulty only if all are true:

  • TimeGate: 3 timed runs completed within time consecutively
  • TransferGate: pass 3 variants (same rubric)
  • VarianceGate: stability rating improves over 2 weeks
  • OperatorGate: routine runs without tutor prompts

If gates fail:

  • truncate: reduce choice, revert to simpler prompts
  • stitch: rebuild binds + buffers + redo transfer

Copyable Almost-Code Block

EducationOS.Runner.Starter12Weeks.v1.0:
Daily:
MicroOutput6 + (NodeBindExpand or InputHarvest15)
Weekly:
2x TimedRun: ScopeLockQ -> 5BoxPlan -> TimedCoreOutput -> ChecklistSweep
1x Repair: RewriteWeakestPara
1x Transfer: TransferVariant3
1x Calibrate: BacktestCalibrate
Monthly:
FullSimPaper at weeks {4,8,12}
LogSensors(minimal):
TimeUsed, Stability(1-5), Transfer(P/F), TopFailureMode, RepairLatency
Calibration:
change one variable per week only
ScaleOnlyIf:
TimeGate + TransferGate + VarianceGate + OperatorGate

Start Here: 

Start Here:

eduKateSG Learning Systems: 

Recommended Internal Links (Spine)

Start Here for Lattice Infrastructure Connectors


Start here if you want the full sequence:

Vocabulary OS Series Index:

Fence English Learning System: 

eduKateSG Learning Systems: 

Recommended Internal Links (Spine)

Start Here for Lattice Infrastructure Connectors