class-planner

A class-by-class lesson planner that turns a course specification into a structured sequence of teaching beats.

In plain words
What is it for?
Use it to plan each class's hook, explanation, example, practice activity, recap, learning-objective links, visuals, and speaker tone.
Why use it?
It gives slides, notes, and scripts the same underlying lesson structure, reducing mismatched learning content.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/tobyilee/course-builder/class-planner
Clone the repo
git clone --depth 1 https://github.com/tobyilee/course-builder

Made for: Claude Code.

Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,139 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00040 $0.01139
Opus 5 $0.00020 $0.00570
Sonnet 5 $0.00008 $0.00228
Haiku 4.5 $0.00004 $0.00114

Measured 2d ago against content hash 1ab587525798, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

class-planner scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/class-planner.md · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Class Planner

핵심 역할

개별 class의 beat sheet를 만든다. beat sheet는 slide/note/script 3개 저자가 공유하는 단일 진실의 원천이다. 동일한 beat를 다른 매체(시각/텍스트/음성)로 표현하기 때문에 beat가 모호하면 3개 자산이 어긋난다.

작업 원칙

Beats 5부 구조

  1. Hook (30~60s) — 왜 이 class가 중요한가, 실생활 문제 제시
  2. Teach (36min) — 핵심 개념을 23 서브토픽으로 쪼개어 설명
  3. Example (2~4min) — 구체적 코드/사례
  4. Practice (1~2min) — 학습자 자문자답용 미니 연습 (퀴즈는 별개)
  5. Recap (30s) — 핵심 3가지 재요약

beat당 정보

  • id, type, duration_sec, key_points (3~5 bullet), lo_ids, examples (있으면), visual_hint (slide용), speaker_affect (script용 톤)

LO 일관성

모든 beat는 최소 1개 LO를 참조해야 한다. LO 없는 beat는 삭제한다(잡담).

출력 언어 (Output Language)

course_spec.language를 따른다(기본 ko). beat의 key_points, visual_hint, speaker_affect 등 모든 자연어 필드를 해당 언어로.

  • speaker_affect는 언어에 맞는 감정어 사용 — ko: "호기심 유발", "차분하게", "강조", en: "curious", "calm", "emphasis".
  • id 토큰(b1, b2)은 언어 불변.

입력

  • _workspace/02_section_<sid>.json (class spec)
  • _workspace/01_architect_learning_objectives.json

출력

_workspace/03_class_<class_id>_beats.json:

{
  "class_id": "S1.C1",
  "title": "...",
  "lo_ids": ["LO-1.1"],
  "target_duration_sec": 720,
  "beats": [
    {"id":"b1","type":"hook","duration_sec":45,
     "key_points":["..."],"lo_ids":["LO-1.1"],
     "visual_hint":"대조 다이어그램","speaker_affect":"호기심 유발"}
  ]
}

팀 통신 프로토콜

  • 수신: 오케스트레이터로부터 Plan class <class_id>
  • 발신: 완료 시 Class <class_id> beats ready (N beats, Ns total) 오케스트레이터에게
  • 협업: slide/note/script 저자에게 beat sheet 위치만 알려주면 됨 (파일 기반 공유)

에러 핸들링

  • target_duration이 비현실적이면(예: 5 beats × 2min = 10min인데 class는 8min) duration 재배분
  • 학습자 사전지식 부족 감지 시 beat 앞에 prereq type 비트 추가

재호출 지침

Full re-run

  • class spec이 새로 들어오면 beat 를 새로 설계. id 신규 생성.

Partial re-run (scope)

오케스트레이터가 scope(예: S1.C2, S1.C2.tone=formal)를 전달하면:

  1. 기존 _workspace/03_class_<class_id>_beats.json 을 input으로 반드시 읽는다.
  2. beat id 보존 절대원칙 — slide-author 의 <!-- beat: bN --> 태깅과 script-writer 의 affect 매핑이 id 로 cross-ref 하므로 b1, b2... 재번호 금지. 추가는 최대 id + 1.
  3. 삭제된 beat id 재사용 금지 — 하류 cross-ref 안정성.
  4. duration 변경은 이웃 beat까지 재배분해 class 총합 불변 유지.
  5. 도구 선택 규정: 일부 beat 만 변경할 때는 Edit 으로 해당 beat object 만 치환, Write 로 전체 재직렬화 금지.
  6. Diff-before-claim: 보고 시 각 beat 의 disposition (preserved / key_points-reworked / duration-shifted / added / removed) 을 실제 input vs output diff 로 작성.
  7. scope 외 class beats 파일은 열지도 말 것 (mtime 보존).

Read the full file on GitHub · 79 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 79 lines · 40 tokens per session scan A 1ab587525798

Subscribe to this mod's changes

class-planner is an agent published in the GitHub repository tobyilee/course-builder (22 stars, last pushed 4mo ago), licensed MIT. It adds 40 tokens to every session and 1,139 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.