exam-tutor

Name: exam-tutor
Author: ZeKaiNie

258stars14forks

How to install exam-tutor?

zekainie/universal-examprep-skill/exam-tutor

$npx -y skills add ZeKaiNie/universal-examprep-skill --skill exam-tutor

Installs into the current project.

Summary

按章节惰性加载授课：每次只读当前阶段的一个 wiki 章节，用生活隐喻讲概念、解剖公式；重点题固定走题面图→问题→读图量→公式→演算→答案详解→溯源七步，画图题先运行算法。用于讲懂当前章或老师勾出的重点题。

Files · 1

View on GitHub

SKILL.md

1	# exam-tutor — chapter teaching
2
3	## Purpose
4
5	Teach exactly one current wiki chapter, using metaphors and formula dissection. In zero-basic mode, explain every linked key question with the fixed seven-step walkthrough. Run algorithms before rendering diagrams. This skill teaches; `exam-quiz` alone quizzes and scores.
6
7	## Activation
8
9	Use when `exam-cram` routes the current phase to teaching, or the student asks to learn the current chapter, derive a formula, or explain a key question.
10
11	## Inputs
12
13	- In `processing_mode=lightweight`: one schema-3 visually accepted current-page batch
14	from `.lightweight/session.json` plus its original pages and declared-scope
15	prompt/answer component assets; no compiled wiki is required.
16	- `references/wiki/chN_*.md`: the one current chapter; never read the whole wiki.
17	- `references/teaching_examples.json`: optional examples, read only through the chapter-filtering CLI below; never an answer source.
18	- `study_state.json`: progress source of truth when present; otherwise the generated `study_progress.md` compatibility view.
19
20	## Workflow
21
22	1. Load one slice. Read `study_state.json.processing_mode` first. In
23	`lightweight`, call `lightweight_session.py status`, plan only the current
24	source/page range if it is not already planned, visually inspect those pages,
25	and import the generic item/component manifest with `record-visual`; teach only a
26	schema-3 `visual_ready` batch. A schema-2 `visual_ready` receipt is quarantined
27	read-only: auditably `abandon` it and plan a new attempt, never teach from or
28	silently upgrade it. While still planned, keep `register-answer-dependency`
29	additive; use `set-answer-dependency --reason` to replace/narrow exact answer
30	pages and `remove-answer-dependency --reason` to remove them. Do not call
31	ingestion/OCR, preload later pages, or require a
32	wiki. In `full`, read exactly one current `references/wiki/chN_*.md`. A missing
33	full-mode file means abstain, name it, and never improvise. If full-mode teaching
34	examples exist, run `python "${CLAUDE_SKILL_DIR}/scripts/list_teaching_examples.py"
35	--workspace <ws> --chapter <N> --json` and use only its returned slice. When the
36	full-mode effective cadence below is `step_by_step`, use `--next-pending` instead
37	of loading the whole chapter example slice. A nonzero
38	exit is an invalid/unreadable inventory, not “no examples”; report it.
39
40	2. Teach reproducibly. Give each concept one concrete metaphor. For STEM, state every formula symbol and unit, then one small hand-computable example. Persist math as `$...$` or `$$...$$`; never leave raw `\frac`, `\sum`, or other TeX as the final reading view.
41
42	3. Use every walkthrough block in order for every stored/teacher-flagged question and every linked question in zero-basic mode.
43
44	Full-mode pacing: read the stored preference plus its reported effective and
45	dormant state. `study_state.json.preferences.interaction_style` stores only
46	`batch\|step_by_step`; missing legacy state means `batch`. This optional preference
47	is independent from `processing_mode`, `artifact_mode`, and
48	`answer_explanation_mode`, and is not a fourth required startup choice. Persist an
49	explicit change only with `update_progress.py --workspace <ws> set
50	--interaction-style <batch\|step_by_step>` (or the strictly validated canonical
51	`--pref interaction_style=...`). It never changes the lightweight page-batch route.
52
53	This option applies only to full-mode `teaching_examples.json` items. It does not
54	claim coverage of the chapter bank, typed question units, or the lightweight
55	page-batch route.
56
57	- Effective `batch`: use the normal full-mode flow. A true
58	`preferences.no_questions=true` or any non-full processing mode makes a stored
59	`step_by_step` choice dormant without overwriting it. A stored `batch` choice
60	remains ordinary batch cadence.
61	- Effective `step_by_step`: call `list_teaching_examples.py --workspace <ws>
62	--chapter <N> --next-pending --json`. It requires `processing_mode=full`,
63	`no_questions=false`, exact `current_phase`, and valid scoped manifest/state data.
64	It reads the manifest, state, notebook bindings, and baseline within one
65	consistent workspace lock, then returns the first manifest-ordered pending item.
66	A missing manifest, malformed state, or nonzero selector exit blocks the pacing
67	decision; report it and do not guess another item. Two bindings may not share one
68	`notebook_ref`. Only a missing notebook entry or anchor/marker/hash/revision drift
69	may return to pending with bounded stable diagnostics. Link/reparse topology,
70	non-directory/non-regular targets, path escape, invalid UTF-8, an unterminated
71	fence, parse/block corruption, schema/scope/baseline damage, duplicate evidence,
72	and `unex

People who liked this also viewed

6 picks

Type

exam-tutor

258stars14forks

View on GitHub

How to install exam-tutor?

zekainie/universal-examprep-skill/exam-tutor

$npx -y skills add ZeKaiNie/universal-examprep-skill --skill exam-tutor

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# exam-tutor — chapter teaching
2
3	## Purpose
4
5	Teach exactly one current wiki chapter, using metaphors and formula dissection. In zero-basic mode, explain every linked key question with the fixed seven-step walkthrough. Run algorithms before rendering diagrams. This skill teaches; `exam-quiz` alone quizzes and scores.
6
7	## Activation
8
9	Use when `exam-cram` routes the current phase to teaching, or the student asks to learn the current chapter, derive a formula, or explain a key question.
10
11	## Inputs
12
13	- In `processing_mode=lightweight`: one schema-3 visually accepted current-page batch
14	from `.lightweight/session.json` plus its original pages and declared-scope
15	prompt/answer component assets; no compiled wiki is required.
16	- `references/wiki/chN_*.md`: the one current chapter; never read the whole wiki.
17	- `references/teaching_examples.json`: optional examples, read only through the chapter-filtering CLI below; never an answer source.
18	- `study_state.json`: progress source of truth when present; otherwise the generated `study_progress.md` compatibility view.
19
20	## Workflow
21
22	1. Load one slice. Read `study_state.json.processing_mode` first. In
23	`lightweight`, call `lightweight_session.py status`, plan only the current
24	source/page range if it is not already planned, visually inspect those pages,
25	and import the generic item/component manifest with `record-visual`; teach only a
26	schema-3 `visual_ready` batch. A schema-2 `visual_ready` receipt is quarantined
27	read-only: auditably `abandon` it and plan a new attempt, never teach from or
28	silently upgrade it. While still planned, keep `register-answer-dependency`
29	additive; use `set-answer-dependency --reason` to replace/narrow exact answer
30	pages and `remove-answer-dependency --reason` to remove them. Do not call
31	ingestion/OCR, preload later pages, or require a
32	wiki. In `full`, read exactly one current `references/wiki/chN_*.md`. A missing
33	full-mode file means abstain, name it, and never improvise. If full-mode teaching
34	examples exist, run `python "${CLAUDE_SKILL_DIR}/scripts/list_teaching_examples.py"
35	--workspace <ws> --chapter <N> --json` and use only its returned slice. When the
36	full-mode effective cadence below is `step_by_step`, use `--next-pending` instead
37	of loading the whole chapter example slice. A nonzero
38	exit is an invalid/unreadable inventory, not “no examples”; report it.
39
40	2. Teach reproducibly. Give each concept one concrete metaphor. For STEM, state every formula symbol and unit, then one small hand-computable example. Persist math as `$...$` or `$$...$$`; never leave raw `\frac`, `\sum`, or other TeX as the final reading view.
41
42	3. Use every walkthrough block in order for every stored/teacher-flagged question and every linked question in zero-basic mode.
43
44	Full-mode pacing: read the stored preference plus its reported effective and
45	dormant state. `study_state.json.preferences.interaction_style` stores only
46	`batch\|step_by_step`; missing legacy state means `batch`. This optional preference
47	is independent from `processing_mode`, `artifact_mode`, and
48	`answer_explanation_mode`, and is not a fourth required startup choice. Persist an
49	explicit change only with `update_progress.py --workspace <ws> set
50	--interaction-style <batch\|step_by_step>` (or the strictly validated canonical
51	`--pref interaction_style=...`). It never changes the lightweight page-batch route.
52
53	This option applies only to full-mode `teaching_examples.json` items. It does not
54	claim coverage of the chapter bank, typed question units, or the lightweight
55	page-batch route.
56
57	- Effective `batch`: use the normal full-mode flow. A true
58	`preferences.no_questions=true` or any non-full processing mode makes a stored
59	`step_by_step` choice dormant without overwriting it. A stored `batch` choice
60	remains ordinary batch cadence.
61	- Effective `step_by_step`: call `list_teaching_examples.py --workspace <ws>
62	--chapter <N> --next-pending --json`. It requires `processing_mode=full`,
63	`no_questions=false`, exact `current_phase`, and valid scoped manifest/state data.
64	It reads the manifest, state, notebook bindings, and baseline within one
65	consistent workspace lock, then returns the first manifest-ordered pending item.
66	A missing manifest, malformed state, or nonzero selector exit blocks the pacing
67	decision; report it and do not guess another item. Two bindings may not share one
68	`notebook_ref`. Only a missing notebook entry or anchor/marker/hash/revision drift
69	may return to pending with bounded stable diagnostics. Link/reparse topology,
70	non-directory/non-regular targets, path escape, invalid UTF-8, an unterminated
71	fence, parse/block corruption, schema/scope/baseline damage, duplicate evidence,
72	and `unex

People who liked this also viewed

6 picks

Type