Section Formats

Formats define how content is delivered — the presentation layer. Each section in a course has both a type (what it means) and a format (how it looks and behaves).

The Enabled flag controls which formats are available to the generator. Disabled formats remain in the system but must not appear in course outlines or content plans.

Overview

Text formats

Text

Status: ✅ Enabled

Meaning and when to use: A foundational format focusing on written information without additional visual elements. Use when clarity and precision are paramount and visual elements are not necessary. Ideal for detailed explanations, definitions, or instructions.

Instructions:

• Structure content for clarity. Use a mix of paragraphs, bullet points, and numbered lists depending on the material. Not every point requires a bullet; use them where they naturally enhance understanding.
• Rather than always posing a reflective question at the end, consider other engaging strategies: a brief real-world application, a challenge for the learner to relate the concept to their own experiences, or a fun fact related to the topic.
• Continuity is essential. Ensure the content is fresh and does not echo what has already been said in other sections. Be especially mindful of adjacent sections to avoid repetition.
• A recap or summary is valuable but not mandatory for every section. Use it if the section is dense or introduces several key points.
• Maintain a professional yet engaging tone. Avoid starting with generic introductions like "Welcome to..." unless contextually appropriate.
• Format with HTML:
  • Prioritize readability: consistent hierarchy with headers and subheaders. Use <strong> for bold, <em> for italics, <ul> or <ol> for lists.
  • If tables are needed, use <table>, <tr>, <th>, <td>.
  • Add emojis where appropriate to enhance content. Use sparingly.
  • NEVER start the section with the topic title — it will be displayed separately.

Output format: Non-strict HTML. Generate text content with flexible HTML markup (e.g., <p>, <ul>, <h4>). Content will be rendered as-is. For lists, Remix icons with Line-style may be used.

Example output:

<p>Start your learning with a clear focus.</p>
<ul>
  <li>Define your goals.</li>
  <li>Stay present in meetings.</li>
</ul>
<ol>
  <li>Step one</li>
  <li>Step two</li>
  <li>Step three</li>
</ol>

VideoPresentation

Status: ✅ Enabled

Meaning and when to use: A structured, visually appealing format based on slides. Use for organized, memorable content that highlights key points. Best for introductions, summaries, dos and don'ts, comparisons, quotes, or definitions. Not suitable for scenario-based narration.

Instructions:

• The content consists of structured material presented on slides. First create a text version of the script.
• Follow the same structural guidelines as Text format.
• Format with HTML (same rules as Text).
• NEVER start the section with the topic title — it will be displayed separately.

Output format: Non-strict HTML. Generate text content with flexible HTML markup (e.g., <p>, <ul>, <h4>). Content will be rendered as-is.

Example output:

<h4>Self-management: a clear definition</h4>
<p>Self-management means choosing and directing your attention, energy, and actions so your work matches your meaningful goals.</p>
<h4>What you can manage</h4>
<ul>
  <li><strong>Attention:</strong> What you choose to focus on right now.</li>
  <li><strong>Energy:</strong> Your mental and physical capacity.</li>
  <li><strong>Actions:</strong> The next step you take.</li>
</ul>

VideoTalkingHead

Status: ✅ Enabled

Meaning and when to use: A dynamic format featuring a person speaking directly to camera. Use to create a personal connection or explain concepts conversationally. Effective for lectures, interviews, or when a human touch is needed.

Instructions:

• Each scene should be in a separate paragraph with detailed narration.
• Provide only the script without any other explanations.
• Begin with a captivating opening: a thought-provoking question, a surprising fact, or a brief overview of what will be covered.
• Ensure logical flow from one point to the next. The script should tell a coherent story.
• Avoid mentioning the video length.
• Never start the script with "Welcome to..." unless this is the first section of the first topic.

Output format: Non-strict HTML. Generate text content with flexible HTML markup (e.g., <p>, <h4>). Content will be rendered as-is.

Example output:

<p>Hello everyone. Today we are discussing the importance of hydration and the benefits it brings to our overall health.</p>
<p>Water makes up around 60% of our bodies and plays a vital role in various bodily functions — regulating body temperature, lubricating joints, aiding digestion, and carrying nutrients to cells.</p>

VideoScenario

Status: ⛔ Disabled

Meaning and when to use: Intended to illustrate specific examples of human behavior through narrative storytelling — explaining causes of a problem, behavioral patterns, or social interactions. Ideal for use cases, mini cases, personal touch, and sections with possible narration.

⚠️ Disabled. Do not use in course outlines or content plans. When VideoScenario is re-enabled, include exactly 3 scenario sections in the course outline.

Instructions:

• Come up with an initial idea: define the central theme, tone, target emotion, and overall storyline.
• Define key plot beats (exposition, conflict, resolution), characters (from the narrative character list), and setting.
• Include mandatory phrases and details that must be said to maintain educational value.
• Include narrator lines as voice-over if necessary.
• The plot description should serve as a comprehensive prompt-instruction for another specialist to create the video — include as many clear and precise requirements as possible.

Output format: Non-strict HTML. Generate text content with flexible HTML markup (e.g., <p>, <h4>). Content will be rendered as-is.

Text + Image formats

TextWithPhoto

Status: ✅ Enabled

Meaning and when to use: Combines text with one realistic image to add authenticity. Use to show real examples, create an emotional connection, or make content more relatable. Suitable for success stories, case studies, or atmospheric content. Can depict narrative characters.

Instructions:

• Structure content for clarity using paragraphs, bullet points, or numbered lists.
• Include one relevant image in a realistic style that visually reinforces the topic.
• The image must have an alt attribute describing the scene. Avoid verbs like "generate" or "search" in the alt text.
• Include the names of narrative characters in the image alt if they appear in the context.
• Maintain a professional yet engaging tone.

Output format: Non-strict HTML for text, with one <img> tag. The image must include an alt attribute with instructions for generation. Set src as "placeholder.jpg". For lists, Remix icons with Line-style may be used.

Example output:

<p>Boost your focus with simple mindfulness techniques.</p>
<ul>
  <li>Define your goals.</li>
  <li>Stay present in meetings.</li>
</ul>
<img src="placeholder.jpg" alt="Two professional colleagues (Alex and Diana) sitting at a conference table, leaning in toward each other in animated conversation, with laptops open and coffee cups nearby">
<p>Start your journey with a clear focus.</p>

TextWithIllustration

Status: ✅ Enabled

Meaning and when to use: Pairs text with custom visuals to simplify or abstract ideas. Use to visualize complex concepts, maintain a specific style, or explain abstract ideas. Best for models, frameworks, or comparisons.

Instructions:

• Structure content for clarity.
• Include one illustration that visually represents what is described in the section.
• The image must have an alt attribute describing the illustration (no "generate" or "search" verbs).

Output format: Non-strict HTML for text, with one <img> tag. Set src as "placeholder.jpg". For lists, Remix icons with Line-style may be used.

Example output:

<p>Boost your focus with simple mindfulness techniques.</p>
<img src="placeholder.jpg" alt="Stylized illustration of a person holding a large flashlight, illuminating a bright icon of a goal, while dark cartoonish worry clouds are left in the shadows.">
<ul>
  <li>Define your goals.</li>
  <li>Stay present in meetings.</li>
</ul>

TextWithPortrait

Status: ✅ Enabled

Meaning and when to use: Pairs text with an illustration of a famous or well-known person mentioned in the text. Use when presenting a concept and its originator or authors. Suitable for historical contexts, research studies, and real examples.

Instructions:

• Structure content for clarity.
• Include one image of the character mentioned in the content.
• The alt attribute must include the person's name and the scene. No "generate" or "search" verbs.
• Avoid character overacting — no feigned emotions.

Output format: Non-strict HTML for text, with one <img> tag. Set src as "placeholder.jpg". For lists, Remix icons with Line-style may be used.

Example output:

<p>Boost your focus with simple mindfulness techniques.</p>
<img src="placeholder.jpg" alt="Portrait of Dr. Robert Chilindi sitting in his study with books on the table.">
<ul>
  <li>Define your goals.</li>
  <li>Stay present in meetings.</li>
</ul>

TextWithDiagram

Status: ✅ Enabled

Meaning and when to use: Integrates text with diagrams to visualize systems or processes. Use to explain structures, sequences, relationships, or processes. Effective for illustrating how things work or are organized.

Instructions:

• Structure content for clarity.
• Include a diagram description to visually represent the concept. The description should be detailed and clear for rendering but must not include actual diagram markup (no SVG or canvas).
• Place the diagram description in a <div class="diagram-description"> tag, describing elements like shapes, labels, and connections.
• Avoid large amounts of text inside the diagram; use only necessary labels.

Output format: Non-strict HTML for text, with one <img> tag containing a detailed description of the diagram in the alt attribute. For lists, Remix icons with Line-style may be used.

Example output:

<p>Understand the mindfulness process to stay calm.</p>
<ul>
  <li>Define your goals.</li>
  <li>Stay present in meetings.</li>
</ul>
<img src="placeholder.jpg" alt="Flowchart with three steps: 1) Pause (circle, labeled Pause), 2) Breathe (rectangle, labeled Take 3 Breaths), 3) Proceed (circle, labeled Act with Focus). Steps connected with arrows.">

Assessment formats

ChoiceQuestions

Status: ✅ Enabled

Meaning and when to use: An interactive assessment format with single or multiple correct answers. Each block contains 1–4 questions on the same topic.

Use for:

• Testing factual knowledge and recall
• Checking understanding and interpretation
• True/False and multiple-choice questions
• Reinforcing learning and encouraging attention to detail

Not for: Self-reflection, self-awareness, self-assessment, or opinion-based questions. This format is for knowledge verification only.

Instructions:

• Keep all questions on one topic.
• Include 1–4 questions.
• Avoid obvious answers — mix up questions and answer options.
• Each question has a type: SingleChoiceQuestion or MultipleChoiceQuestion.
• Include helpful feedback for each question explaining what is correct.
• SingleChoiceQuestion must have only ONE correct answer.
• Use basic HTML tags. Do NOT use Markdown.

Output format: JSON. Wrap questions in an array.

Each question structure:

• type — SingleChoiceQuestion | MultiChoiceQuestion
• question — string
• possible_answers — array of { text: string, correct: boolean }

SingleChoiceQuestion has feedback_text in each answer option. MultiChoiceQuestion has correct_feedback_text and incorrect_feedback_text at question level.

Example output:

{
  "questions": [
    {
      "type": "SingleChoiceQuestion",
      "question": "<p>You need the finance department to approve a budget increase. Which <strong>option</strong> fits best for this audience?</p>",
      "possible_answers": [
        { "text": "A detailed report with charts showing financial return.", "correct": true, "feedback_text": "<p>Correct! Financial stakeholders respond to data.</p>" },
        { "text": "A presentation focused on improving team morale.", "correct": false, "feedback_text": "<p>Not quite. Finance prioritizes ROI, not morale.</p>" },
        { "text": "An email citing a senior leader's preference.", "correct": false, "feedback_text": "<p>Let's try again. Authority alone doesn't persuade financial reviewers.</p>" }
      ]
    },
    {
      "type": "MultiChoiceQuestion",
      "question": "<p>What <strong>increases</strong> your chance of approval for a new product idea?</p>",
      "possible_answers": [
        { "text": "Show a short customer interview clip.", "correct": true },
        { "text": "Provide a quantified impact estimate.", "correct": true },
        { "text": "Add unrelated inspirational quotes.", "correct": false },
        { "text": "Ignore open risks to keep the mood positive.", "correct": false }
      ],
      "correct_feedback_text": "<p>Correct. Evidence and quantified impact build trust.</p>",
      "incorrect_feedback_text": "<p>Review what makes a business case credible.</p>"
    }
  ]
}

FillTheBlanks

Status: ✅ Enabled

Meaning and when to use: An interactive format where learners complete a text by filling in missing words or terms. Blanks are shuffled to prevent memorization.

Use for:

• Testing recall and terminology accuracy
• Strengthening memory retention and comprehension
• Definitions, formulas, or step-by-step processes
• Contextual application in instructions or case studies

Not for: Self-reflection, self-awareness, self-assessment, or open-ended discussions. This format is for knowledge verification only.

Instructions:

• 3–6 blank placeholders per task.
• Prefer to use only dropdowns or only inputs (do not mix unless necessary).
• Ensure blanks are meaningful and test specific knowledge. Shuffle — do not follow the original order of the explanation.
• For dropdowns: include at least 1 distractor per dropdown. Shuffle options; avoid placing the correct option first.
• For inputs: define 1 correct answer.
• Include positive feedback for correct answers and hints for incorrect ones.
• Avoid overly complex sentences.
• Do not use textarea. This format is not for free answers.
• Do not use Markdown. Basic HTML tags are acceptable.
• Available placeholders: [PLACEHOLDER_INPUT] or [PLACEHOLDER_DROPDOWN].

Output format: JSON with fields:

• type — string, required
• content — string with placeholders, required
• placeholders — array in the same order as in content, required
• correct_feedback_text — string, required
• incorrect_feedback_text — string, required

Input placeholder: { type, index, options: [{ text, correct: true }] } Dropdown placeholder: { type, index, options: [{ text, correct: boolean }] } (shuffled)

Example output:

{
  "content": "<p>The capital of <strong>France</strong> is [PLACEHOLDER_INPUT]. It is famous for the [PLACEHOLDER_DROPDOWN] Tower and lies in [PLACEHOLDER_DROPDOWN].</p>",
  "placeholders": [
    { "index": 1, "type": "input", "options": [{ "text": "Paris", "correct": true }] },
    { "index": 2, "type": "dropdown", "options": [
      { "text": "Eiffel", "correct": true },
      { "text": "Colosseum", "correct": false },
      { "text": "Big Ben", "correct": false },
      { "text": "Brandenburg", "correct": false }
    ]},
    { "index": 3, "type": "dropdown", "options": [
      { "text": "Asia", "correct": false },
      { "text": "Africa", "correct": false },
      { "text": "Europe", "correct": true },
      { "text": "Oceania", "correct": false }
    ]}
  ],
  "correct_feedback_text": "Correct: Paris, Eiffel, Europe. Well done!",
  "incorrect_feedback_text": "Review European capitals, landmarks, and continents, then try again."
}

Question

Status: ⛔ Disabled

Meaning and when to use: A simple free-text field without correctness validation. Use when learners should share reflections, opinions, or examples rather than demonstrate knowledge. Suitable for reflection questions and self-assessment.

⚠️ Disabled by design. No pedagogical advantage over QuestionWithFeedback. Do not use in course outlines or content plans. Can be re-enabled via the Enabled flag.

Instructions:

• Formulate a clear question or task with a free-text answer.
• No correctness check is performed.
• Suitable for reflection exercises, brainstorming, or personal input.
• Basic semantic HTML is acceptable.

Output format: JSON. Wrap questions in an array. Each question: { question: string }.

QuestionWithFeedback

Status: ✅ Enabled

Meaning and when to use: An open-ended assessment format with an input box and LLM-evaluated criteria. A separate LLM uses the criteria to review and provide feedback.

Use for:

• Assessing depth of understanding, reasoning, and analysis
• Evaluating critical thinking and creative problem-solving
• Diagnosing learner preferences or style based on their answer

Suitable for both Self_Assessment and Knowledge_Assessment.

Instructions:

• Formulate ONE clear question or task.
• Avoid complex specific technical tasks.
• Include a criteria string with a model answer or clear evaluation criteria.
• Make criteria specific but not revealing the exact answer.
• Important: Criteria should be a clear, self-contained prompt for another LLM with isolated context that will give feedback. Include all necessary keywords.
• The question should indicate what is expected in the answer.
• Basic semantic HTML is acceptable: <ul>, <ol>, <p>, <strong>.

Output format: JSON with fields:

• content — iframe string pointing to https://labs.studytube.ai/question-with-feedback with URL params: question, criteria, blueprint, language. Include section title in iframe title attribute.
• question_data.question — string
• question_data.criteria — string

Example output:

{
  "content": "<iframe src=\"https://labs.studytube.ai/question-with-feedback?question=[question]&criteria=[criteria]&blueprint=[blueprint]&language=[language]\" title=\"[section title]\"></iframe>",
  "question_data": {
    "question": "<p>What have you learned about strategy, tactics, and results?</p>",
    "criteria": "The answer should include at least 3 points covering strategy, tactics, and results."
  }
}

Artifact

Status: ✅ Enabled

Meaning and when to use: An AI-driven interactive format that goes beyond static assessments. Enables hands-on learning across recognition, application, critique, and synthesis by aligning the interaction pattern, cognitive level, and learning goal. Supports custom quizzes, games, simulations, flashcards, and structured interactive tasks.

Use an Artifact only when ALL are true:

1. It reinforces a specific learning objective already taught in the module.
2. The learner must perform a meaningful cognitive action (recognize, differentiate, apply, analyze/evaluate, or synthesize).
3. The interaction enables immediate, structured checking — not open-text AI grading.

Allowed use cases:

• Active recall and practice
• Applying a taught model or rule to examples
• Gamified drills and lightweight scenario checks
• Diagnostic self-assessment (no right/wrong, but structured output)

Do not use Artifact for:

• Long or open-text answers evaluated by AI → use QuestionWithFeedback
• Introducing new knowledge not taught earlier
• Tasks with no clear advantage over static questions
• Multi-step scenarios
• Image-based hotspot tasks
• Video-analysis tasks

Instructions (outline stage — blueprint): Return a concise blueprint with exactly these fields:

• Learning Outcome: After completing this artifact, the learner should be able to…
• Cognitive Level: Recognition | Differentiation | Application | Analysis + Evaluation | Synthesis
• Interaction Summary: Describe only the learner's mental action (e.g., categorize, sequence, apply, diagnose, compare). Do not mention UI controls.
• Content Scope: Specify only previously taught knowledge. No new material.

Constraints:

• One clear learning outcome only.
• One primary cognitive action only.
• Do not describe UI elements (no drag-and-drop, dropdowns, sliders, etc.).
• Do not mention template or layout names.

Instructions (content stage — game data): Create a data-driven concept that can be directly used to generate the game configuration.

Required output (Markdown):

• Title — sentence case, reflects the essence of the task and the section title
• Assessment type — knowledge_assessment or self_assessment
• Learning outcome map — target outcome, cognitive level, what exactly is being assessed
• Data for the gaming experience — explicit list of all items including distractors, exact elements that MUST be included (verbatim), correctness rules and feedback rules, canonical correct order/set
• Permissible variations — what can vary (word order, shuffling), what cannot vary (terminology, number of elements, canonical sequence)

Output format: Markdown