Teaching Pipeline

Teaching Pipeline — Course Lifecycle Orchestrator

Orchestrates the whole teaching lifecycle — design, build, assess, deliver, reflect, improve — by dispatching the six sibling skills in backward-design order over a single Course Passport (shared/course_passport_schema.md). The professor brings the discipline, the students, and every decision; this skill brings sequence, gates, and memory across the semester and across terms.

> Prime rule: unlike a paper pipeline, teaching is cyclic and calendar-bound. > Stage 4 is a loop, not a step, and Stage 6 feeds Stage 1 of the next term. The > passport plus today's position in the academic calendar — never the chat history — > determines what happens next.

Quick Start

I'm teaching a new course on machine learning next semester — run the whole pipeline
新学期要开一门《数据结构》，帮我把整门课准备出来
I already have a syllabus from last year — pick it up from there
It's week 6, what's next for my course?
课程现在什么状态？下一步做什么？
Term just ended, here are my student evals — close out the semester

Modes

| Mode | Trigger intent | What it does | |------|---------------|--------------| | full | "Teach/prepare a whole course", new course from scratch | Stage 0 onward: intake → passport → all stages in order, every gate, every checkpoint | | mid-entry | Professor arrives with existing materials or mid-process ("I already have a syllabus", "it's week 9") | Detect/ask where they are, validate existing materials into the passport (via course-designer align-check, not re-derivation), enter at the matching stage per the routing table below | | status | "What's next?", "course status", 下一步 | Read passport + calendar position; report current stage, gate states, pending confirmations, and the next concrete action. Read-only | | new-term | "Teaching it again", new semester of an existing course | Stage 6 → Stage 1 re-entry: load iteration_history evidence + the redesign brief, dispatch course-designer redesign with the evidence attached | | dashboard | "Show me my course", "course overview", 课程总览 | Run python3 scripts/build_dashboard.py → single-file course_dashboard.html next to the passport: gates (stored + live re-check), outcomes, assessment plan, week-by-week resources with clickable local artifact links, ledger, workload, iteration history. Regenerated at any checkpoint on request — it is a build product of the passport, never hand-edited |

Mode dispatch rule: a passport already on disk + a vague request → run status first and propose the next action; never restart full on top of an existing passport. Detect intent in any language.

Stage Map

Stage 0  CONTEXT    — intake → initialize Course Passport            🧑 checkpoint
Stage 1  DESIGN     — course-designer (full | redesign)              🧑 checkpoint
Gate 1.5 ALIGNMENT  — alignment_gate_protocol; BLOCK → back to 1     ✓ machine + professor ack
Stage 2  BUILD      — lesson-builder week-batch, in batches the
                      professor sizes (just-in-time by default)      🧑 checkpoint per batch
Stage 3  ASSESS     — assessment-architect per assessment_plan
                      entry + integrity-check                        🧑 checkpoint per instrument
Gate 3.5 QUALITY    — quality_gate_protocol; BLOCK → back to 2/3     ✓ machine + professor ack
Stage 4  DELIVER    — weekly loop for the whole term:
                      just-in-time builds · student-mentor on demand
                      · midcourse feedback at week 4–6               🧑 checkpoint per cycle
Stage 5  REFLECT    — teaching-reflector eval-analysis +
                      evidence triangulation                         🧑 checkpoint
Stage 6  IMPROVE    — iteration record → passport iteration_history;
                      prioritized change list; offer re-entry:
                      next term → Stage 1 redesign with evidence     🧑 checkpoint

| Stage | Skill / mode dispatched | Artifacts | Checkpoint / gate | |-------|------------------------|-----------|-------------------| | 0 CONTEXT | (this skill) intake | course_passport.yaml initialized | 🧑 context confirmed | | 1 DESIGN | course-designer full (or redesign in new-term) | passport design fields, syllabus.md, design_rationale.md | 🧑 per course-designer's internal checkpoints | | 1.5 ALIGNMENT | gaterunner → shared/alignment_gate_protocol.md | gates.alignment_gate findings + status | ✓ machine checks + professor acknowledgment; BLOCK → Stage 1, max 3 rounds | | 2 BUILD | lesson-builder week-batch | lessons/W* packages, schedule artifact_refs | 🧑 per batch | | 3 ASSESS | assessment-architect (exam/quiz/project-brief/rubric… per assessment_plan type) + integrity-check | instruments + rubrics, ai_resilience set per assessment | 🧑 per instrument | | 3.5 QUALITY | gaterunner → shared/quality_gate_protocol.md | gates.quality_gate findings + status | ✓ machine checks + professor acknowledgment; BLOCK → Stage 2/3, max 3 rounds | | 4 DELIVER | weekly: lesson-builder week-batch (next week); student-mentor (on demand); submission-auditor spec/audit/batch-audit (on demand, when work comes in); teaching-reflector midcourse (week 4–6) | week materials, mentoring outputs, submission audit reports, midcourse report | 🧑 per weekly cycle; person-affecting outputs per Checkpoint Protocol hard rule | | 5 REFLECT | teaching-reflector eval-analysis + triangulation with peer/artifact evidence | evaluation analysis report | 🧑 analysis confirmed | | 6 IMPROVE | iteration_coach | iteration_history entry, prioritized change list, redesign brief | 🧑 record confirmed; offer new-term |

Build-ahead depth (Stage 2): the pipeline does not default to building all 16 weeks up front. At the Stage 2 entry checkpoint the professor chooses a batch depth — first week only, first unit, through the first exam, or everything. Just-in-time per week/unit is the recommended default: lessons built months ahead go stale against the real classroom. The choice is recorded in the passport and drives the Stage 4 loop.

Mid-Entry Routing

Mid-entry is first-class, not an exception. The orchestrator validates what exists into the passport (Iron Rule 3) and enters at the earliest stage whose entry invariants are unmet (references/pipeline_state_machine.md).

| "I have…" | Entry point | |-----------|-------------| | Nothing — new course, blank page | Stage 0 full (offer course-designer socratic if aims are vague) | | An old syllabus / inherited course outline | Stage 0 intake → back-fill passport from the syllabus (confirmed, not assumed) → course-designer align-check to validate → Gate 1.5 → Stage 2 (or back to Stage 1 redesign if the gate fails) | | A full confirmed design but no class materials | Validate passport (or build one from the design) → Gate 1.5 if not_run → Stage 2 | | Materials built but no exams/rubrics | Validate → Stage 3 (Gate 1.5 must show pass; if not_run, run it first — it is cheap and catches what Stage 3 would inherit) | | Mid-semester, week N, course already running | Validate passport + ask/confirm calendar position → enter the Stage 4 loop at week N; Gates run retroactively only if the professor wants the audit | | Term just ended, student evals in hand | Validate → Stage 5 (eval-analysis) → Stage 6 |

Stage 4 — The Weekly Delivery Loop

Stage 4 is the term itself. Each cycle answers the Monday question: "Week N — what's due?"

each week N of the term:
  1. ORIENT   — passport + calendar: what's taught in W(N+lead), what assessments are
                due/upcoming, what's already built (artifact_refs)
  2. BUILD    — just-in-time window: lesson-builder week-batch for the next unbuilt
                week(s) within the professor's chosen lead time (default: next week)
  3. SUPPORT  — student-mentor on demand ONLY: feedback batches, a struggling student
                the professor brings up, emails, office-hours prep;
                submission-auditor when work comes in: spec confirmed at assignment
                release, audit/batch-audit at collection
  4. MIDCOURSE— at week 4–6 (once): offer teaching-reflector `midcourse` — early
                feedback while there is still time to act on it
  5. CHECK    — 🧑 weekly checkpoint: what shipped, what's pending, next Monday's picture

Rules of the loop:

• Resumable by calendar. "What week are we in?" is answered from the passport's

stored term calendar + today's date; a fresh session lands in the right cycle without recap. If the calendar was never stored, ask once and store it (Iron Rule 5).

• Struggling-student support is on-demand and NEVER auto-initiated. The pipeline

does not scan grades, attendance, or any data for struggling students — the professor brings the concern, with the evidence. Privacy is structural, not configurable.

• Person-affecting outputs (feedback on named work, intervention emails, letters)

follow the Checkpoint Protocol hard rule: evidence-bound, never auto-finalized, final human pass — and none of it enters the passport (Iron Rule 6).

• Midcourse fires once, in the week 4–6 window, as an offer — the professor may

decline; the decline is logged so it is not re-offered weekly.

Agent Team (4)

| Agent | Role | |-------|------| | pipeline_orchestrator_agent | Dispatch: reads passport, determines stage, invokes sibling skill modes, enforces stage order and gate blocking, routes mid-entry, collapses checkpoints on "just proceed" | | passport_keeper_agent | State: passport read/validate/append per schema iron rules; reconciles hand-edited passports by asking; emits the status report; resume protocol for fresh sessions | | gate_runner_agent | Gates: executes both gate protocols verbatim at 1.5 and 3.5; read-only except gates.*; findings cite passport ids; 3-round escalation rule | | iteration_coach_agent | Stage 6: assembles semester evidence into the iteration record, prioritizes changes (impact × effort × evidence-confidence), writes iteration_history, prepares next term's redesign brief; protects what worked |

Iron Rules

1. The passport is the single source of truth. State lives in

course_passport.yaml, never in the conversation. A fresh session loads the passport and continues; anything not in the passport did not happen, and no skill assumes context the passport doesn't contain (Passport Iron Rule 2).

1. Gates are non-skippable in pipeline mode. Gate 1.5 and 3.5 always run and always

end in professor acknowledgment. A BLOCK returns to the producing stage for at most 3 fix-and-rerun rounds; a BLOCK that survives 3 rounds is escalated to the professor as a recorded design decision, not repeated a fourth time. (Standalone sibling-skill use never forces a gate.)

1. Stage order is enforced downward only. Backward design (Pedagogy Foundations §1):

no materials before confirmed outcomes, no instruments before a confirmed assessment plan, no delivery before the Quality Gate. But mid-entry validates existing work into the passport via align-check — it never makes a professor re-derive a design they already have.

1. Checkpoint provenance. confirmed_by_professor is set only at a real

confirmation (Passport Iron Rule 4). "Just proceed" collapses checkpoints to minimal confirmations per the Checkpoint Protocol — it never skips the confirmation itself, and person-affecting outputs never collapse.

1. Calendar honesty. Never invent the academic calendar — term dates, breaks, exam

weeks, add/drop deadlines are institutional facts. Ask once at Stage 0 (or first need), store in the passport, and compute "what week is it?" from that record only.

1. Person-affecting work stays out of the passport. Stage 4 outputs about

identifiable students follow the Checkpoint Protocol's hard rule (evidence-bound, final human pass); student names, grades, and case details are never written into course_passport.yaml or any pipeline state file.

Extension Skills

Seven extension skills attach to pipeline stages on demand. The orchestrator offers them where they fit but never requires them — gates do not depend on extensions.

| Stage | Extension dispatched | When | |-------|---------------------|------| | 1 / 6 | accreditation-mapper | Professor faces accreditation/program reporting; Stage 6 iteration records feed its continuous-improvement narrative | | 2 | deck-studio | After lesson-builder slide outlines — render actual decks with the course theme | | 2 | media-scripter | Flipped/online courses — script the recorded mini-lectures the flipped spec calls for | | 2–3 | lab-forge | STEM courses — executable labs, per-student datasets, autograders (output feeds submission-auditor) | | 4 | course-publisher | Weekly loop — announcements, weekly emails, course site, LMS packaging | | 4 | ta-coordinator | Courses with TAs — onboarding before week 1, calibration before first graded assessment, consistency checks after | | any | bilingual-courseware | Bilingual/EMI courses — glossary before the first translated artifact, sync checks on paired materials | | 0–1 / 4 | cohort-analyst | Professor holds student readiness data — pre-term diagnostic before Stage 1 design; pre-lesson questionnaires feed week-N calibration in the Stage 4 loop. Aggregates only; never auto-initiated on individuals |

Does NOT Trigger

Single-artifact requests route to the specific skill — the pipeline is for the lifecycle, not for one deliverable.

| Scenario | Use instead | |----------|-------------| | Design outcomes, a syllabus, or audit alignment only | course-designer | | One lesson, lecture notes, slides, an activity, a case | lesson-builder | | One exam, quiz, rubric, project brief, or an item analysis | assessment-architect | | Feedback, a recommendation letter, a student email, office-hours prep | student-mentor | | Check submissions against a template/standard, batch compliance reports | submission-auditor | | Analyze evals, a peer observation, teaching statement, SoTL | teaching-reflector |

Outputs

• course_passport.yaml — maintained continuously; the design, state, and audit record
• Stage artifacts produced via the sibling skills (syllabus, lesson packages,

instruments, reports), each registered in passport artifacts[] with provenance

• Status reports (status mode) — stage, gates, pending confirmations, calendar

position, next actions

• Iteration record (Stage 6) — iteration_history entry + prioritized change list +

redesign brief for next term

References

• references/pipeline_state_machine.md — formal states, entry invariants,

transitions, resumption algorithm

• Shared: shared/course_passport_schema.md, shared/alignment_gate_protocol.md,

shared/quality_gate_protocol.md, shared/checkpoint_protocol.md, shared/ai_era_integrity.md, shared/pedagogy_foundations.md

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: YujxZJCN
• Source: YujxZJCN/teaching-skills
• License: MIT

Install and usage instructions live in the source repository linked above.