工作台会话挂载了教师课堂录像/讲义材料时使用。
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Extract how the teacher teaches, then teach the user's requested topic in that style. The source material is evidence about delivery and presentation; it is not automatically factual source material for the new course.
The learner-facing name for this mode is 「名师风格」. Never expose the words "clone" or "style transfer" to the learner. The result should simply feel like a lesson taught in the evidenced style.
Call list_materials before drawing conclusions. Identify the attached source
materials and confirm extraction is complete. For recordings, verify that the
session exposes the transcript and keyframe derivatives. If extraction is idle,
call extract_material first; if it is pending or running, call
wait_for_materials and list again; if it failed, tell the user what failed instead
of inventing a profile from incomplete evidence.
Use read_material to page through every transcript derivative from offset 0
to the final page. Do not sample only the beginning. As you read, build a
compact style profile in your reasoning covering:
Support each claim with short transcript evidence and its timestamp. Frequency and distribution matter: distinguish a repeated habit from a phrase used once.
When the transcript contains a [keyframe@mm:ss](mat_xxx) marker, sample the
referenced frame with read_material. Use frames across the recording rather
than adjacent duplicates. Record visual evidence such as board-writing density,
slide composition, hierarchy, color use, teacher position, gestures, and the
relationship between speech and what remains on screen.
Do not infer visual style when the recording has no readable keyframes.
Use search_material to test specific transcript hypotheses after the full
read. Search catchphrases and recurring sentence forms, compare their frequency,
and revisit the surrounding snippets. Search verifies a proposed pattern; it
never replaces reading the transcript in order.
Treat unsupported or contradicted hypotheses as absent from the final profile.
Plan the course in the conversation, then create_stage and call
generate_scene once per page with an explicit brief. Put the complete,
evidence-backed style profile into each page's brief under a ## 讲课风格
section, alongside any independently verified subject facts the requested
course needs. Include the load-bearing instructions: opening ritual, phrasing
habits, example pattern, pacing, interaction rhythm, and visual presentation.
For every page, put the relevant evidence and concrete style directions in
generate_scene.materialFacts. Examples: 「以复习上节课的问题开场」、「先用一个具体例子建立
直觉，再给出公式」、「页尾用已核验的口头禅式小结」。Do not merely
refer to a profile by name: the page generator only receives the text placed
in each page's brief and materialFacts.
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Evidence-based creation of a reusable personal course-making Skill from the user's own classroom and chat history. Use when the user asks to summarize, analyze, or learn from their past course-making records/history and create a personal, exclusive, or reusable Skill, such as “总结我的做课记录，给我做个专属 Skill” or “从我过去的课程和对话里提炼一个个人技能”.
日本語の概要は準備中です。原文の説明を表示しています。
Multi-stage series — a request for several classrooms that belong together, like 「7 天学 Python」, a four-week onboarding track, a semester unit split across lessons, or "turn this book into a lesson per chapter". Clarifies the series brief, gets the user's explicit sign-off on the full lesson list before spending anything, then builds the stages one at a time into a shared folder. Use when the ask is a sequence of classrooms; not for a single stage, however large.
日本語の概要は準備中です。原文の説明を表示しています。
Deep Interactive courses. Plans the course so the learner manipulates something on most pages — simulations, diagrams they explore, code they run, games, 3D scenes — with slides reserved for the opening frame and the closing consolidation. Use when the topic has a mechanism, a variable to vary, a process to trace, or a structure to inspect.
日本語の概要は準備中です。原文の説明を表示しています。
Courses whose content rests on current, external or real-world facts that must be verified against live sources before being taught — recent events, market or policy data, product and version specifics, scientific developments, named cases. Researches the topic first, keeps a claim-to-source ledger, and grounds the outline and every page in what was actually fetched. Use when the request depends on up-to-date or externally checkable facts; not for timeless textbook topics that stand on established knowledge alone.
日本語の概要は準備中です。原文の説明を表示しています。
Improve factual reliability while creating or reviewing a course or supplied content. Use when the user asks to fact-check, verify accuracy, reduce hallucinations, make a reliable course, or mentions 事实性错误、知识性错误、专业知识准确性、可靠性. During creation, checks the completed pages before delivery; on existing content, returns a short evidence-backed report and lets the user choose what to fix. Not for grammar, style, or layout. Combine with deep-research when current evidence is the course's main subject.
日本語の概要は準備中です。原文の説明を表示しています。
Turn any concept, lesson, slide deck, or source material into a Feynman learning cycle in which learners explain first, expose the smallest gap, rebuild the explanation through Socratic prompts, strip jargon, stress-test analogies, and transfer the idea to a new context. Use when the user asks for teach-back, learning by explaining, or a Feynman-style classroom. Do not use when the Feynman technique itself is merely the lesson topic or when study strategy should remain a parallel goal.
日本語の概要は準備中です。原文の説明を表示しています。