# From Prompt to Production: activity prompts Fifty activities. Select a small set for live practice; the rest form a reusable activity library. Estimated activity times exclude transitions and discussion. Images are accompanied here by real text. Starter-kit inputs are at the top level of faculty-starter-kit/. Replace bracketed context with your own course details. ## 01. From Prompt to Production **START WITH A NEED · About 2 minutes** Pick one course material you keep rebuilding. Bring that task through this workshop. **Make:** One real teaching need **Check:** What would make it better? **Tool:** Any chat or agent environment **Inputs:** One recurring teaching task, or the supplied campus sustainability brief. ```text Help me choose one course-development task for a short production sprint. My recurring task is [describe it]. My learners are [audience]. Ask no more than two necessary questions, then suggest a small deliverable, the evidence that would make it usable, and what still requires my judgment. Keep the scope small enough to review during one session. Save my choice as drafts/teaching-need.md. ``` **Steps** 1. Name one thing you rebuild each term. 2. Choose one small artifact rather than a whole course. 3. Write down what would count as a useful improvement. **Review** - The task has a real audience and purpose. - The deliverable is small enough to inspect. **Expected output:** One real teaching need **Sources** Workshop activity design; no external factual claim required. ## 02. Define the win **START WITH A NEED · About 3 minutes** Describe the learners, the learning outcome, and the evidence you want to see. **Make:** A three-line design brief **Check:** Could another instructor use it? **Tool:** Any chat or agent environment **Inputs:** A teaching need and the COURSE_BRIEF.md example. ```text Turn this need into a three-line brief: learners and prior knowledge; one observable learning outcome; evidence students will produce. Example: first-year sustainability students compare two campus transport proposals and justify a recommendation using supplied data and one stated limitation. Flag missing context without inventing course policy. Save drafts/design-brief.md. Offer one stronger and one weaker example of evidence so I can compare. ``` **Steps** 1. Describe learners and prior knowledge. 2. Replace 'understand' with an observable performance. 3. Decide what student work will reveal that performance. **Review** - The outcome describes student work. - The evidence fits the outcome. - No policy or course fact was invented. **Expected output:** A three-line design brief **Sources** Workshop activity design; no external factual claim required. ## 03. Give the agent a brief **START WITH A NEED · About 5 minutes** Write the course context, approved sources, constraints, and review criteria in one reusable file. **Make:** COURSE_BRIEF.md **Check:** Are the requirements specific? **Tool:** Cursor, Antigravity, or a text editor **Inputs:** faculty-starter-kit/COURSE_BRIEF.md and the example inputs. ```text Read COURSE_BRIEF.md. Propose improvements to this reusable course-development brief. Include learners, prior knowledge, outcomes, teaching context, approved inputs, boundaries, output format, and review criteria. Preserve facts in the supplied files, mark unknowns as questions, and do not silently fill them in. Save a proposed revision as drafts/COURSE_BRIEF-proposed.md, leaving the original intact. ``` **Steps** 1. Open the supplied brief and replace the example context with your course. 2. Add approved readings and explicit constraints. 3. Ask the agent to read the file at the start of each task. **Review** - Brief states audience, outcome, sources, deliverable, and review criteria. - Missing facts remain visible. - Agent can summarize the requirements accurately. **Expected output:** COURSE_BRIEF.md **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart ## 04. Build a course workspace **START WITH A NEED · About 5 minutes** Open a dedicated folder in Cursor or Antigravity. Add your brief, sources, drafts, and approved materials. **Make:** A reusable course folder **Check:** Can you find the current version? **Tool:** Cursor or Antigravity **Inputs:** A copy of faculty-starter-kit. ```text Inspect this course workspace. Read COURSE_BRIEF.md and the file list. Explain which inputs are approved teaching materials, which are fictional samples, and which are synthetic data. Propose a simple organization using sources/, drafts/, and release/. Create only missing folders; do not move or overwrite source files. Write drafts/workspace-guide.md identifying the input and output locations. ``` **Steps** 1. Copy the starter kit into a new folder you control. 2. In Cursor open the folder; in Antigravity create a project and add the folder. 3. Ask the agent to list and summarize the files before it builds. **Review** - Agent is operating in the intended folder. - It recognizes synthetic and fictional inputs. - Original inputs remain unchanged. **Expected output:** A reusable course folder **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Antigravity Getting Started: https://www.antigravity.google/docs/getting-started ## 05. Cursor: plan first **START WITH A NEED · About 6 minutes** Ask the agent to plan one small resource from your brief. Review the plan before asking it to build. **Make:** An editable build plan **Check:** Does every step serve the outcome? **Tool:** Cursor Plan Mode **Inputs:** COURSE_BRIEF.md and APPROVED_READING.md. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Plan a five-minute campus transport learning activity as a self-contained local HTML file. Include one outcome, a short explanation, a decision using supplied evidence, and a reflection. First propose the content, files, and acceptance checks in drafts/build-plan.md. Do not implement yet. Identify unavailable dependencies and prefer a single HTML file with no external libraries. ``` **Steps** 1. Choose Plan Mode and submit the prompt. 2. Read and edit the plan: scope, proposed files, and acceptance checks. 3. Choose Build once the plan matches the goal; inspect the resulting file. **Review** - The plan names a testable output. - The activity actually elicits the outcome. - The build has no unnecessary accounts or dependencies. **Expected output:** An editable build plan **Sources** - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode ## 06. Antigravity: review the work **START WITH A NEED · About 6 minutes** Give the agent a bounded task. Inspect its plan, files, and preview before accepting the result. **Make:** A reviewed first draft **Check:** What evidence shows it works? **Tool:** Antigravity Planning with artifact review **Inputs:** A local workshop project, COURSE_BRIEF.md, and APPROVED_READING.md. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Plan a short student activity that compares two campus transport proposals. Produce an implementation plan with the student instructions, expected response, files, and verification steps. Wait for my review before implementation. After approval, build drafts/transport-activity.html and show a preview or walkthrough. State which checks you actually completed and which require me to test. ``` **Steps** 1. Create a project and add the workshop folder; use Local Mode. 2. Use Planning and a Request Review policy for a review pause. 3. Comment on the plan, then inspect the built files and walk through the preview. **Review** - Plan incorporates the course brief. - Files and preview match the request. - Reported tests include any checks not completed. **Expected output:** A reviewed first draft **Sources** - Antigravity Getting Started: https://www.antigravity.google/docs/getting-started - Antigravity Artifact Review: https://www.antigravity.google/docs/artifact-review ## 07. Give each role a job **START WITH A NEED · About 6 minutes** Assign a researcher, writer, and critic different tasks. Give each role an output and a review criterion. **Make:** A three-role workflow **Check:** Who checks the evidence? **Tool:** Agent roles; subagents if available **Inputs:** The course brief, approved reading, and one draft. ```text Use three explicitly assigned roles to improve this activity. Researcher: map each factual claim to a passage in APPROVED_READING.md and record limitations in drafts/source-ledger.md. Writer: create drafts/activity-v1.md from the brief and supported claims. Critic: compare that draft against the outcome, evidence, and learner context; write concrete issues to drafts/critique.md. Keep the outputs separate. Do not invent sources. I will review the findings before revision. ``` **Steps** 1. Give all roles the same brief and approved sources. 2. Assign a separate output to each role. 3. Compare the critic's findings with the evidence yourself. **Review** - Role outputs have distinct purposes. - Critique points to specific passages or requirements. - Instructor checks unresolved disagreements. **Expected output:** A three-role workflow **Sources** - Cursor Subagents: https://cursor.com/docs/subagents ## 08. Critique, then revise **START WITH A NEED · About 5 minutes** Ask a critic to find three specific weaknesses. Revise the draft and explain what changed. **Make:** A stronger second version **Check:** Can you see the improvement? **Tool:** Any agent environment **Inputs:** drafts/activity-v1.md plus the brief and critique; create a first draft if needed. ```text Read the course brief and drafts/activity-v1.md. Identify the three weaknesses that most affect learning or usability. For each, quote the relevant portion, explain the effect on learners, and propose a precise revision. Then produce drafts/activity-v2.md and drafts/change-log.md after I approve the changes. Preserve supported factual content and list any unresolved uncertainty. ``` **Steps** 1. Compare the draft against three explicit criteria. 2. Choose changes that matter to learning. 3. Inspect version one and version two side by side. **Review** - Every change addresses a stated issue. - The outcome is preserved. - Revision has not introduced new unsupported claims. **Expected output:** A stronger second version **Sources** Workshop activity design; no external factual claim required. ## 09. Generate more than one idea **DESIGN FOR LEARNING · About 4 minutes** Generate ten ways to teach one difficult concept. Vary the medium, participation, and classroom setting. **Make:** An idea menu **Check:** Are the ideas meaningfully different? **Tool:** Any chat or agent environment **Inputs:** One difficult concept, such as correlation versus causation. ```text Generate ten substantially different activities for teaching correlation versus causation to first-year students. Include individual, pair, small-group, online, and classroom options. For each provide the student action, evidence of learning, approximate time, needed materials, and one access consideration. Avoid ten variations of a quiz. Use fictional examples and identify them as such in the instructor materials. Save drafts/idea-menu.md. ``` **Steps** 1. Start with one concept and a clear learner level. 2. Ask for variation in student actions and settings. 3. Circle two ideas worth combining. **Review** - Ideas vary in the work students do. - Each includes observable evidence. - Materials and time are realistic. **Expected output:** An idea menu **Sources** Workshop activity design; no external factual claim required. ## 10. Choose, combine, refine **DESIGN FOR LEARNING · About 5 minutes** Compare three ideas against learning value, preparation time, and access. Combine the strongest parts into one activity. **Make:** A teachable activity concept **Check:** Can you run it next week? **Tool:** Any chat or agent environment **Inputs:** Three selected ideas from the idea menu. ```text Compare these three activity ideas for a 20-minute first-year class segment. Use learning value, preparation effort, participation barriers, and the evidence students produce. Explain tradeoffs instead of inventing numerical precision. Combine the strongest compatible elements into one activity and save drafts/activity-concept.md. Include a no-projector fallback and a one-sentence reason for your choice. ``` **Steps** 1. Compare the same criteria across three ideas. 2. Select a workable combination. 3. Test the combined activity against real classroom time. **Review** - The selection follows stated priorities. - Combination does not overload the time available. - There is a usable low-tech version. **Expected output:** A teachable activity concept **Sources** Workshop activity design; no external factual claim required. ## 11. Use Bloom's with evidence **DESIGN FOR LEARNING · About 6 minutes** Turn a vague outcome into observable student work. Try an explanation, an analysis, and a justified design. **Make:** Three outcome-and-task pairs **Check:** What will students actually do? **Tool:** Any chat or agent environment **Inputs:** A vague outcome: 'Students understand campus sustainability.' ```text Replace 'Students understand campus sustainability' with three different outcome-and-task pairs: explain a tradeoff, analyze two proposals using evidence, and design a justified recommendation. Use the revised Bloom framework to name the cognitive process and relevant knowledge dimension for each. State the observable student product and criteria for success. Do not claim the verb alone establishes complexity or that Create is always the best choice. Save drafts/bloom-alignment.md. ``` **Steps** 1. Name the student performance and the evidence it produces. 2. Compare explanation, analysis, and design tasks. 3. Match the level to your actual course purpose. **Review** - Classification considers the whole task. - Evidence demonstrates the stated process. - The task fits learners' prior knowledge. **Expected output:** Three outcome-and-task pairs **Sources** - Krathwohl: A Revision of Bloom's Taxonomy: https://doi.org/10.1207/s15430421tip4104_2 ## 12. Check the alignment **DESIGN FOR LEARNING · About 5 minutes** Give the agent an outcome, an activity, and an assessment. Ask where they match and where they drift. **Make:** An alignment map **Check:** Does the assessment test the outcome? **Tool:** Any chat or agent environment **Inputs:** One outcome, classroom activity, and assessment. ```text Audit this trio: outcome—justify a campus transport recommendation using supplied evidence; activity—compare two proposals and discuss tradeoffs; assessment—list three definitions. Map the cognitive work and evidence in each. Identify the mismatch, then propose two assessment revisions that fit the outcome while retaining a feasible marking load. Save drafts/alignment-audit.md with a before-and-after example. ``` **Steps** 1. Give the agent a deliberately misaligned example. 2. Ask it to explain the mismatch. 3. Revise the assessment or the stated outcome deliberately. **Review** - Assessment samples the intended performance. - Suggested revisions are feasible. - Supporting prerequisite knowledge remains available. **Expected output:** An alignment map **Sources** - Krathwohl: A Revision of Bloom's Taxonomy: https://doi.org/10.1207/s15430421tip4104_2 ## 13. UDL: invite engagement **DESIGN FOR LEARNING · About 5 minutes** Redesign one activity with a meaningful choice, a clear purpose, and a low-stakes way to begin. **Make:** An engagement redesign **Check:** Does the choice preserve the goal? **Tool:** Any chat or agent environment **Inputs:** One class activity, its outcome, and likely participation barriers. ```text Review this campus transport activity through the UDL principle of engagement. Identify likely barriers in relevance, participation, and getting started. Redesign it with a meaningful choice of context, a clear explanation of purpose, and a private low-stakes first response before group discussion. Preserve the shared outcome and criteria. Save drafts/engagement-redesign.md with a short reason for each change. ``` **Steps** 1. Identify a barrier in the activity rather than label a learner. 2. Offer a meaningful choice that serves the outcome. 3. Include an easy, low-stakes first step. **Review** - Choices remain relevant to the same outcome. - Students can begin without public exposure. - The redesign reduces an identified barrier. **Expected output:** An engagement redesign **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ ## 14. UDL: offer another route **DESIGN FOR LEARNING · About 6 minutes** Explain one difficult concept with concise text, an accessible visual, and a concrete example. **Make:** Three ways into one concept **Check:** Is the meaning consistent? **Tool:** Cursor Agent, Antigravity, or an approved chat tool **Inputs:** COURSE_BRIEF.md and one instructor-approved explanation. Use the campus sustainability brief, or supply a short ecology explanation that learners find difficult. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Choose one important concept that first-year students need to understand, and state that concept before proceeding. Build three equivalent explanations: (1) a plain-language paragraph of no more than 100 words; (2) a diagram specification with a concise text alternative and an adjacent explanation of the essential relationships; and (3) a concrete campus example with one transfer question. Keep the same central meaning and any qualifications in all three. Use only information in the reading; mark anything that needs instructor input. In drafts/representation-options.md, give the three versions, the barrier each may reduce, and a side-by-side check for changes in meaning. Do not categorize students into fixed learning styles. If you cannot access the files, ask me to paste them. ``` **Steps** 1. Identify the central concept and the barrier: unfamiliar terminology, a hidden relationship, or lack of a familiar context. 2. Ask for the three representations and inspect one sentence that carries the central idea across all three. 3. Compare the diagram description to the source; revise any visual implication that overstates causality. 4. Choose the combination that supports this lesson and save the reviewed version. **Review** - All versions preserve the same claim, scope, and uncertainty. - The diagram can be understood with its adjacent text explanation. - The concrete example clarifies the concept rather than replacing it with a different idea. - No fixed learning-style labels are used. **Expected output:** drafts/representation-options.md with three explanations of one concept, text alternatives, and a meaning-consistency check. **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ ## 15. UDL: vary the response **DESIGN FOR LEARNING · About 7 minutes** Design two ways students can demonstrate the same outcome. Use shared criteria and identify essential requirements. **Make:** Equivalent response options **Check:** Are the expectations comparable? **Tool:** Cursor Agent, Antigravity, or an approved chat tool **Inputs:** COURSE_BRIEF.md and APPROVED_READING.md; preserve the full outcome, including comparison, two pieces of evidence, a tradeoff, and a limitation. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Design two response options for this outcome: 'Compare two campus transport proposals and justify a recommendation using at least two relevant pieces of supplied evidence, one tradeoff, and one limitation.' Option A is a short written recommendation. Option B is a narrated diagram with a text transcript. Specify a comparable scope, audience, evidence requirement, and expected effort for each; do not assume matching word counts guarantees equivalence. Create a shared criteria table for accuracy, use of evidence, reasoning, and treatment of the tradeoff and limitation. Distinguish essential outcome requirements from optional presentation choices. Explain whether either option would be inappropriate if the actual course outcome explicitly assesses academic writing or oral delivery. Save the instructions and common criteria to drafts/response-options.md. Flag conflicts with the course brief for instructor decision. ``` **Steps** 1. State exactly what is being assessed and which communication skills, if any, are essential. 2. Generate the two response options and a single criteria table. 3. Check whether each option can reveal the same reasoning and evidence use. 4. Adjust scope and required supports, then explain the options to a colleague as if they were a student. **Review** - Both options expose the intended reasoning and evidence. - Criteria do not reward irrelevant production polish. - Essential disciplinary or communication requirements remain explicit. - Students have clear instructions and sufficient access to the required tools. - Both formats require two relevant evidence details, comparison, a tradeoff, and a limitation. **Expected output:** drafts/response-options.md containing two student-facing submission routes and shared criteria. **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ ## 16. Build tomorrow's lesson **DESIGN FOR LEARNING · About 8 minutes** Turn the chosen activity into a lesson with an opening question, practice, feedback, and an exit ticket. **Make:** A ready-to-review lesson plan **Check:** Where will you see learning? **Tool:** Cursor Agent or Antigravity with the course folder open **Inputs:** COURSE_BRIEF.md, APPROVED_READING.md, and an activity selected earlier in the workshop. ```text Use COURSE_BRIEF.md and APPROVED_READING.md to turn the selected campus transport decision activity into a 20-minute lesson for the audience in the brief. Create drafts/lesson-plan.md with four phases: an opening question that surfaces an assumption, guided practice using the supplied evidence, feedback that helps learners revise their reasoning, and a two-minute exit ticket. For each phase include time, what the instructor does, what students do, the evidence of learning, and the materials needed. The timings must total 20 minutes. Include one anticipated misconception and a response, plus a low-technology alternative. Keep instructor notes separate from the student handout. Do not add factual claims or numerical data not supported by the approved reading. After drafting, check alignment between the outcome, practice, feedback, and exit ticket; report any mismatch before revising. ``` **Steps** 1. Select one outcome and one activity rather than asking for a whole course. 2. Generate the timed lesson and read only the student actions first. 3. Check that practice and the exit ticket require the intended cognitive work. 4. Revise one mismatch and save a student handout separately from the facilitator plan. **Review** - The timings total the planned class time. - Each activity has a clear student action and observable evidence. - The exit ticket assesses the intended outcome rather than general enjoyment. - The plan works with the stated class size and available materials. **Expected output:** drafts/lesson-plan.md and a short student handout, ready for instructor review. **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf ## 17. Sharpen the question **RESEARCH WITH EVIDENCE · About 6 minutes** Turn a broad topic into three researchable questions. Compare scope, key terms, and the evidence each would need. **Make:** A focused research question **Check:** Could you actually investigate it? **Tool:** An approved chat tool, Cursor Agent, or Antigravity **Inputs:** A broad faculty-selected research topic and a realistic project constraint. Example: formative feedback in first-year writing, explored through a focused literature review. ```text Act as a research-design thinking partner. My broad topic is formative feedback in first-year university writing, and my immediate task is a focused literature review, not a new experiment. Propose three different researchable questions. For each, identify the population or context, the key concept or relationship, the likely evidence needed, useful search terms, and a reason the scope might be too broad. Separate questions about effectiveness from questions about students' experiences; do not imply the same study design answers both. Do not invent citations, findings, or claims of a literature gap. Recommend one question that is feasible for a small teaching-focused review and explain what I would need to narrow or define. Save drafts/research-questions.md if file writing is available. ``` **Steps** 1. Choose a topic that matters to the instructor's teaching rather than starting from a tool feature. 2. Generate three questions with different scopes or evidence needs. 3. Compare which question could be investigated with the available time and access. 4. Select one question and rewrite it in language a colleague can interpret unambiguously. **Review** - The question names a context and a concept that can be investigated. - The evidence needed could plausibly answer the question. - The wording does not assume the result. - Scope matches the available time and source access. **Expected output:** drafts/research-questions.md with three candidates, an explicit scope, and one selected question. **Sources** - ACRL Framework for Information Literacy for Higher Education — Research as Inquiry: https://www.ala.org/acrl/standards/ilframework ## 18. Design the search **RESEARCH WITH EVIDENCE · About 8 minutes** Ask for keywords, synonyms, database choices, and inclusion criteria. Test the search in a library database. **Make:** A search strategy **Check:** What does the search miss? **Tool:** Cursor or Antigravity for drafting; an institutional library database for testing **Inputs:** The selected question from drafts/research-questions.md and access to a relevant library database. ```text Read drafts/research-questions.md and design a transparent first search for the selected question. For the example topic formative feedback in first-year writing, propose concept groups, synonyms, spelling variants, and candidate subject headings that I must verify in the database thesaurus. Show one broad Boolean search and one narrower alternative, using parentheses so the logic is clear. Suggest two relevant database types or named databases and explain why they fit; do not claim we searched them. Propose inclusion and exclusion criteria, and explain the effect of any date or language limit instead of imposing arbitrary limits. Create drafts/search-strategy.md and an empty drafts/search-log.csv with columns database, search_date, exact_query, filters, result_count, changes, and rationale. I will run the query and enter actual results. Do not invent counts or papers. ``` **Steps** 1. Review the concept groups; remove terms that prematurely exclude relevant literature. 2. Run the exact query in a library database and record the date and results. 3. Inspect a few clearly relevant and irrelevant results to diagnose the search. 4. Revise one term, subject heading, or limit and document why. **Review** - Every reported result count is observed rather than generated. - Boolean groups express the intended question. - The search retrieves at least some known relevant material when available. - Limits and omitted terminology are visible. **Expected output:** drafts/search-strategy.md and drafts/search-log.csv containing at least one genuinely executed search. **Sources** - ACRL Framework for Information Literacy for Higher Education — Searching as Strategic Exploration: https://www.ala.org/acrl/standards/ilframework ## 19. Build a source ledger **RESEARCH WITH EVIDENCE · About 10 minutes** Record each source's title, author, year, link, and relevant passage. Open every source before using it. **Make:** A verified source ledger **Check:** Can you trace every claim? **Tool:** Cursor Agent or Antigravity, plus a browser or library reader **Inputs:** Three participant-selected research sources verified from the actual literature. Place authorized copies or links in sources/. The starter kit does not supply research papers. ```text Build a source ledger for the real research materials I have provided in sources/. Use only those materials and their verified publication pages. For each source, record source_id, exact title, author or organization, year, DOI or stable URL, source type, the specific claim relevant to my question, page or section, a short supporting passage or faithful paraphrase, limitations, and verification status. Distinguish 'source exists', 'full text read', and 'claim checked'. If you cannot access a full text or confirm metadata, mark it unverified and state what I need to check; never complete missing bibliographic details from guesswork. Save drafts/source-ledger.csv and a concise drafts/source-checks.md. Do not treat APPROVED_READING.md in the starter kit as a published research study. ``` **Steps** 1. Open each publication or source page and confirm that the source exists. 2. Read the relevant passage in context, checking whether it supports the intended claim. 3. Generate the ledger and inspect one row against the original page. 4. Correct metadata or overstatement, and keep inaccessible sources visibly unverified. **Review** - Titles, authors, years, and identifiers match the source. - The cited passage supports the precise claim and its scope. - Unverified or inaccessible materials remain labeled as such. - Paraphrases do not add causality, certainty, or generality absent from the source. **Expected output:** drafts/source-ledger.csv with traceable claims and an honest verification status for every row. **Sources** - ACRL Framework for Information Literacy for Higher Education: https://www.ala.org/acrl/standards/ilframework ## 20. Compare the evidence **RESEARCH WITH EVIDENCE · About 8 minutes** Use verified readings to build a matrix of claims, methods, contexts, and limitations. Keep disagreement visible. **Make:** An evidence matrix **Check:** Which claims have direct support? **Tool:** Cursor Agent, Antigravity, or an approved document-grounded chat tool **Inputs:** drafts/source-ledger.csv and the verified full texts or authorized excerpts used to build it. ```text Read the verified source ledger and the supplied readings. Create drafts/evidence-matrix.csv with one row per source and columns research_question, main_claim, method, sample_or_context, outcome_or_concept, result, limitation, relevant_page_or_section, and relevance_to_our_question. Use 'not reported' when the source does not provide a field. Then write drafts/evidence-comparison.md: identify agreements, disagreements, and cases that cannot be directly compared. Tie every comparison to source IDs and passages. Separate an author's interpretation from a directly reported result. Do not count repeated references to the same study as independent evidence, do not force different outcomes into a common metric, and do not turn this small source set into a claim about the whole field. ``` **Steps** 1. Verify the extraction from one source before scaling to the others. 2. Compare contexts and methods before comparing conclusions. 3. Ask the agent to identify one apparent disagreement that may reflect different questions or measures. 4. Revise the synthesis so disagreements and incomparable evidence remain visible. **Review** - Extracted claims accurately reflect their source passages. - The matrix represents methods and contexts rather than stripping them away. - Disagreement is explained or preserved, not erased. - Missing information is not filled with guesses. **Expected output:** drafts/evidence-matrix.csv and a short comparison memo tied to the verified source ledger. **Sources** - ACRL Framework for Information Literacy for Higher Education — Scholarship as Conversation: https://www.ala.org/acrl/standards/ilframework ## 21. Find a gap worth testing **RESEARCH WITH EVIDENCE · About 7 minutes** Ask what your source set does not answer. Turn one possible gap into a question and a follow-up search. **Make:** A tentative gap statement **Check:** Is it a gap or a search limit? **Tool:** Cursor Agent or Antigravity, followed by a library search **Inputs:** The verified evidence matrix, selected research question, and actual search log. ```text Read drafts/evidence-matrix.csv and drafts/search-log.csv. Identify three questions this particular source set does not answer. For each, explain whether the absence may reflect our search terms, database coverage, date or language limits, inaccessible texts, or a plausible gap in the literature. Do not say 'no research exists' based on this small set. Choose one tentative gap and write it as: 'Within the sources reviewed so far, we have not found ...; we will test this by ...'. Give a follow-up search using additional terms, one citation-chaining action, and the evidence that would disconfirm the gap. Save drafts/gap-hypothesis.md. The instructor will decide whether the gap remains defensible after the follow-up search. ``` **Steps** 1. Identify an unanswered question within the reviewed source set. 2. Check whether the search process plausibly produced that absence. 3. Run a targeted follow-up search and inspect references or citing work. 4. Update or abandon the gap statement if relevant evidence is found. **Review** - The statement is explicitly limited to the reviewed evidence. - Search limitations are separated from substantive gaps. - The follow-up could actually discover counterevidence. - The question remains worth investigating even if the novelty claim changes. **Expected output:** drafts/gap-hypothesis.md with a qualified statement, follow-up search, and disconfirmation criterion. **Sources** - ACRL Framework for Information Literacy for Higher Education — Research as Inquiry: https://www.ala.org/acrl/standards/ilframework ## 22. Invite a counterargument **RESEARCH WITH EVIDENCE · About 7 minutes** Ask a skeptical reviewer to challenge your interpretation using the same sources. Identify evidence that could change your view. **Make:** A stronger argument **Check:** What would change your mind? **Tool:** An independent agent or separate conversation in Cursor, Antigravity, or another approved tool **Inputs:** A draft interpretation, the verified source ledger, and the same source set used by the author. ```text Act as a skeptical but fair reviewer of drafts/evidence-comparison.md. Use only the same verified sources and drafts/source-ledger.csv. State the author's strongest defensible claim before critiquing it. Identify up to three weaknesses in the interpretation, citing the relevant source IDs and passages. Distinguish an unsupported claim, a plausible alternative explanation, and a limitation already acknowledged by the author. Offer the strongest counterargument the supplied evidence supports, and say what further evidence would change your assessment. Do not invent a disagreement or external source merely to appear critical. Save drafts/counterargument-review.md. Then propose precise revisions, but leave the original draft unchanged until I choose which revisions to accept. ``` **Steps** 1. Provide the reviewer the draft, the criteria, and the same evidence. 2. Read whether the reviewer understood the strongest version of the argument. 3. Check each criticism against the source; reject unsupported criticism as well as unsupported praise. 4. Accept a justified revision and record what changed in the argument. **Review** - The critique represents the original claim fairly. - Each substantive challenge is traceable to evidence. - Alternative explanations are labeled as possibilities when appropriate. - The revision improves accuracy without manufacturing false balance. **Expected output:** drafts/counterargument-review.md and an instructor-selected revision to the interpretation. **Sources** - ACRL Framework for Information Literacy for Higher Education — Scholarship as Conversation: https://www.ala.org/acrl/standards/ilframework - Cursor Subagents: https://cursor.com/docs/subagents ## 23. Make a reading guide **RESEARCH WITH EVIDENCE · About 6 minutes** Use an approved reading to draft guiding questions, key terms, and one misconception check. **Make:** A student reading companion **Check:** Can students answer from the reading? **Tool:** Cursor Agent, Antigravity, or an approved chat tool **Inputs:** APPROVED_READING.md from the starter kit, or a faculty-selected reading approved for use in the tool. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Create a student reading companion for the audience in the course brief. Include a one-sentence purpose, three guiding questions in the order students encounter the relevant information, four key terms with plain-language definitions grounded in the reading, and one misconception check that asks students to justify their answer. The reading is an authored classroom scenario; do not describe it as a published research study. Create drafts/reading-guide.md for students and drafts/reading-guide-key.md for the instructor. In the key, identify the heading or paragraph that supports each answer and mark questions that require inference. Do not add outside facts or answerable-looking questions whose answers are absent from the reading. ``` **Steps** 1. Choose a short approved reading and one purpose for reading it. 2. Generate the student guide and separate instructor key. 3. Answer each question using only the reading; mark where inference is required. 4. Revise any question that rewards guessing or introduces unsupported knowledge. **Review** - Each required answer is supported by the reading or explicitly labeled as inference. - Definitions preserve the disciplinary meaning. - The misconception check exposes reasoning, not just a guessed option. - The student file does not accidentally include the answer key. **Expected output:** drafts/reading-guide.md and a separate drafts/reading-guide-key.md tied to the reading. **Sources** - ACRL Framework for Information Literacy for Higher Education: https://www.ala.org/acrl/standards/ilframework ## 24. Brief the class accurately **RESEARCH WITH EVIDENCE · About 8 minutes** Turn verified notes into a one-page research brief. Separate established findings, uncertainty, and questions for discussion. **Make:** A sourced teaching brief **Check:** Did compression change the meaning? **Tool:** Cursor Agent or Antigravity with the verified research folder **Inputs:** Verified notes, drafts/source-ledger.csv, and drafts/evidence-comparison.md from participant-selected real research. For a quick practice, use the authored scenario but call the output a scenario brief. ```text Use the verified research notes, source ledger, and evidence comparison to write a one-page teaching brief for first-year students. Save drafts/teaching-brief.md. Organize it into: what this reviewed evidence supports; what remains uncertain or context-dependent; and three questions for class discussion. Use plain language while preserving important qualifications, contradictory findings, and the difference between association and causation. Give a compact source list with verified links and attach source IDs to the claims they support. Do not add new facts, sources, or a stronger consensus than the notes establish. Then create drafts/brief-audit.md, comparing every substantive claim in the brief with the supporting passage and identifying any meaning lost through compression. If the inputs are the starter kit's fictional campus scenario rather than real research, title the output 'Scenario brief', describe only what the scenario supplies, and do not invent research findings. ``` **Steps** 1. State the intended audience and the one question the brief should help students discuss. 2. Generate a concise brief with uncertainty and discussion kept distinct. 3. Audit one compressed sentence against the original source passage. 4. Revise overstatement and verify every source link before classroom use. **Review** - Every substantive research claim has direct support. - Uncertainty and disagreement remain visible. - The brief does not imply the small source set is exhaustive. - Source links and metadata have been checked. **Expected output:** drafts/teaching-brief.md plus drafts/brief-audit.md recording source support and compression checks. **Sources** - ACRL Framework for Information Literacy for Higher Education: https://www.ala.org/acrl/standards/ilframework ## 25. Brief the image before making it **MAKE VISUAL MATERIALS · About 5 minutes** Specify the learning purpose, subject, composition, style, and what the visual must get right. **Make:** An image design brief **Check:** What should learners notice first? **Tool:** Cursor, Antigravity, or an approved chat tool to write the brief **Inputs:** A module title, learning purpose, intended image placement, and any instructor-approved visual facts. Example: an introductory ecology module on urban wetlands. ```text Help me brief an image for a first-year ecology module on urban wetlands. The image will introduce the topic in an LMS module header, not function as a scientific data diagram. Write drafts/image-brief.md specifying: audience and learning purpose; what learners should notice first; required subject matter; composition and title-safe space; a suitable visual style; image aspect ratio 16:9; details that must be verified by the instructor; and what to exclude. The intended scene is a city-edge wetland with water, shoreline vegetation, and a visible relationship to the surrounding built environment. Avoid implying that a specific real site or species assemblage has been documented. Request an illustration, not a documentary photograph. Keep any eventual title as editable text outside the image. Include a provisional text alternative based on the intended purpose, to be revised after seeing the actual output. Do not generate the image yet. ``` **Steps** 1. Name the image's teaching job: orientation, explanation, comparison, or evidence. 2. Specify composition and factual requirements before selecting a visual style. 3. Check whether the image needs to be decorative or informative in its final context. 4. Save the brief and identify the one detail the instructor will inspect first. **Review** - The image has an explicit learning purpose. - The visual does not imply unsupported site-specific facts. - Title placement and likely crop are considered. - The text alternative will be finalized from the actual image and context. **Expected output:** drafts/image-brief.md with a concrete purpose, visual constraints, and review criteria. **Sources** - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ ## 26. Create a module image **MAKE VISUAL MATERIALS · About 7 minutes** Use an image-capable tool to make a visual for one module. Leave space for a readable title. **Make:** A module hero image **Check:** Does it support the topic? **Tool:** An approved image-capable tool; Cursor or Antigravity can prepare and organize the brief when image generation is not available in that environment **Inputs:** drafts/image-brief.md from the previous activity and an image-capable account available for the workshop. ```text Create one 16:9 landscape illustration for the urban-wetlands module described in my image brief. Show a calm city-edge wetland with visible water, shoreline plants, and distant buildings, so learners notice the relationship between an ecosystem and the built environment. Use a clear editorial illustration style with restrained color and enough separation to remain readable at LMS-header size. Leave the upper-left third simple for a title that I will add as editable text. Include no text, logos, labels, data, or claims about a specific real place. Do not depict a scientific mechanism that the brief has not established. Save or let me download the selected image as assets/urban-wetlands-hero.png. After I inspect it, help me write assets/urban-wetlands-hero-description.md with an appropriate text alternative for its actual use, a record of the prompt, and one factual or visual detail that still needs instructor review. ``` **Steps** 1. Send the reviewed brief to an image-capable tool and generate a first version. 2. Inspect topic fit, subject relationships, title-safe space, and potential visual errors. 3. Place it in a sample module header to test the real crop and text placement. 4. Save the chosen image and a contextual text alternative; revise one specific issue if needed. **Review** - The image supports the stated module topic. - The final crop preserves the important subject and readable title space. - No unsupported documentary or scientific claim is implied. - The text alternative fits the actual purpose and content of the selected image. **Expected output:** assets/urban-wetlands-hero.png and assets/urban-wetlands-hero-description.md, ready to place in an LMS module after review. **Sources** - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ ## 27. Show how a concept works **MAKE VISUAL MATERIALS · About 7 minutes** Build a labeled diagram from an explanation. Check every relationship, arrow, and label against an approved source. **Make:** A concept diagram **Check:** Could a wrong arrow teach an error? **Tool:** Cursor or Antigravity for the diagram specification; an image tool or editable diagram tool for the visual **Inputs:** An instructor-approved explanation saved as sources/concept.md. Self-contained alternative: use the instructor-defined Claim → Evidence → Reasoning argument structure in the prompt. ```text Create an introductory academic-writing diagram for first-year students. Use this definition as the approved source: a claim states a position; evidence provides relevant observations or source material; reasoning explains how that evidence supports the claim. Represent the relationship without implying that evidence automatically proves a claim. First write drafts/argument-diagram-spec.md with the exact three labels, the meaning and direction of each connector, and a short text alternative. Then create an editable SVG and an HTML preview in drafts/. Use high contrast, large labels, and no decorative arrows. Do not invent disciplinary facts or additional causal links. Add a relationship audit listing each arrow and the sentence that justifies it. Show the diagram and audit for instructor review. ``` **Steps** 1. Select one relationship students often misunderstand and supply a short approved explanation. 2. Read the agent's list of nodes and connectors before accepting the visual. 3. Open the HTML preview; trace each arrow aloud and compare it with the source. 4. Correct a relationship before adjusting colours or polish. Keep the specification with the image. **Review** - Does each node match the approved explanation? - Are arrow direction and type meaningful and justified? - Can a reader understand the relationship from the text alternative? - Is the actual visual legible at teaching size? **Expected output:** drafts/argument-diagram-spec.md, drafts/argument-diagram.svg, drafts/argument-diagram.html, and a relationship audit **Sources** - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ ## 28. Improve it in three passes **MAKE VISUAL MATERIALS · About 8 minutes** Refine the visual for accuracy, then clarity, then style. Change one thing at a time and compare versions. **Make:** An iteration strip **Check:** Which change improved learning? **Tool:** Image-generation tool, with Cursor or Antigravity maintaining the brief and change log **Inputs:** A module image from activity 26 or another draft image; the image brief; instructor feedback on one inaccurate detail. ```text Use the attached wetland module image and its teaching brief. Preserve the subject and intended message: learners should notice the relationship between water, vegetation, and habitat. Keep the original as wetland-v1.png. Make three separate revisions and show each before continuing. Pass 1, accuracy: change only the instructor-identified factual problem; ask for clarification if it is unspecified. Pass 2, clarity: reduce competing background detail so the intended relationship is easier to see, without removing essential information. Pass 3, style: make colour and composition consistent with the course without changing content. Save each version separately and write drafts/visual-change-log.md with the requested change, what actually changed, what stayed stable, and whether another check is needed. Do not label a version accurate solely because it looks realistic. ``` **Steps** 1. Keep the starting image and brief visible. 2. Choose one factual correction, then compare the revised image with the original. 3. Make a clarity pass and a style pass, saving a new file each time. 4. Place the versions side by side; ask which change helps learners answer the intended question. **Review** - Was the requested change actually made? - Did unrelated content drift between versions? - Is the teaching relationship clearer? - Are image prompts and accepted versions saved? **Expected output:** Original plus three versioned images and drafts/visual-change-log.md **Sources** - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ ## 29. Write the text alternative **MAKE VISUAL MATERIALS · About 5 minutes** Describe the visual's teaching purpose and essential information. Add a longer description when the image carries complex meaning. **Make:** Alt text and a description **Check:** Could someone learn without seeing it? **Tool:** Cursor, Antigravity, or a text assistant; instructor review against the actual image **Inputs:** The final diagram/image and the sentence explaining what learners should learn from it. ```text Write text alternatives for the attached learning visual in its course context. The image teaches how a claim, evidence, and reasoning work together in an academic argument. Inspect the actual visible content. Produce a short alt text that identifies its purpose, plus a longer adjacent description that explains all essential labels and relationships in a useful reading order. If the image contains data, include the data in a real table. Do not infer information that is not visible or established by the approved explanation. Do not start with 'image of' unless the medium itself matters. Save drafts/image-text-alternative.md. Then answer: what could a learner do from the description that they could do from the image, and what information still requires instructor clarification? ``` **Steps** 1. State the teaching purpose before asking for a description. 2. Compare the proposed description with the actual image and correct unsupported details. 3. For a complex visual, add a structured long description beside it rather than putting everything into an unwieldy alt attribute. 4. Temporarily hide the image and test whether the description supports the intended task. **Review** - Does the description convey the learning purpose? - Are all essential relationships available in text? - Is anything invented or contradicted by the image? - Is the description placed where learners can access it? **Expected output:** drafts/image-text-alternative.md containing a short alternative, long description if needed, and unresolved questions **Sources** - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ ## 30. Storyboard the lesson slides **MAKE VISUAL MATERIALS · About 6 minutes** Plan six slides around one question. Give each slide a single message, a visual idea, and a teaching move. **Make:** A six-slide storyboard **Check:** Does the sequence tell a story? **Tool:** Cursor, Antigravity, or another text assistant **Inputs:** One learning outcome, a driving question, and an approved reading or instructor explanation. ```text Plan a six-slide micro-lesson for first-year students answering: 'Does a strong association mean that one thing caused another?' Learning outcome: students can identify a plausible alternative explanation and state what additional evidence they would seek. Use this instructor-authored fictional example: across twelve weeks, ice-cream sales and pool visits rise together as weather becomes warmer. This example illustrates a possible common cause; it does not establish a measured causal result. Create drafts/correlation-storyboard.md with one row per slide: single message, maximum 25 words of visible copy, visual idea, teaching move, speaker explanation, and source or instructor-example status. Sequence: opening prediction, observed pattern, possible explanation, counterexample, student practice, exit question. Include a short text alternative for each meaningful visual. Do not build the deck until I have reviewed the story. ``` **Steps** 1. Define the question students should be able to answer. 2. Read just the six single-message lines in order; check that the argument develops. 3. Check where students predict, practise, and explain, not only listen. 4. Approve or revise the storyboard before asking for visual production. **Review** - Is the driving question answered by the sequence? - Does every slide advance one message? - Do learners do something with the concept? - Are examples and sources accurately described? **Expected output:** drafts/correlation-storyboard.md with six slide plans and speaker explanations **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ ## 31. Build the slides, then inspect **MAKE VISUAL MATERIALS · About 10 minutes** Ask for a slide file or browser-based deck from the storyboard. Open it, present it, and check readability. **Make:** A reviewable slide deck **Check:** Can the back row read it? **Tool:** Cursor or Antigravity; a browser for preview; presentation software if PPTX generation is available **Inputs:** The approved six-slide storyboard from activity 30. ```text Read drafts/correlation-storyboard.md and build the approved micro-lesson. First state which output formats the available toolchain can actually create. Prefer an editable PowerPoint if supported; otherwise create a self-contained browser deck in drafts/correlation-deck.html with six 16:9 slides, real selectable text, previous/next controls, keyboard navigation, and a printable reading view. Use one clear message per slide, large type, strong contrast, and the approved visual descriptions. Include speaker notes and text alternatives. Do not add unverified facts. Open the actual output, walk through all six slides, and inspect text clipping, reading order, navigation, and the final exit question. Report what you checked and what you could not check. Save a review log; do not say the deck is accessible based only on a screenshot. ``` **Steps** 1. Ask the agent to name the actual file format it can produce. 2. Build from the approved storyboard and open the resulting file. 3. Present it at full screen and view a distant or reduced-size preview to find unreadable text. 4. Test every navigation control and read the companion text; request one targeted revision. **Review** - Does the file open and contain exactly the approved sequence? - Is text readable without zooming? - Are there clipping or contrast problems? - Do keyboard controls and reading view work? - Are the visible claims supported? **Expected output:** An actual six-slide PPTX or drafts/correlation-deck.html, speaker notes, and drafts/deck-review.md **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Google Antigravity Artifact Review: https://www.antigravity.google/docs/artifact-review - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 32. Turn a concept into a short script **MAKE VISUAL MATERIALS · About 6 minutes** Draft a ninety-second explanation with a concrete example, visual cues, and a transcript. Read it aloud and revise. **Make:** A short teaching script **Check:** Does it sound natural when spoken? **Tool:** Cursor, Antigravity, or a text assistant; optional recording tool **Inputs:** A concept explanation and one concrete example; a microphone is optional. ```text Write a ninety-second teaching script for first-year students explaining why association alone does not establish causation. Use the fictional ice-cream-sales and pool-visits example, with warmer weather as a plausible common cause. Begin with a question, use conversational language, and end with a transfer question about another association. Target approximately 180–210 spoken words, but treat the actual read-aloud timing as the test. Output a two-column storyboard with spoken words and visual cues, then a clean transcript without production directions. Describe essential visual meaning in the narration. Save drafts/association-script.md and drafts/association-transcript.md. Flag any phrase that overstates the example or suggests that causal conclusions are never possible. ``` **Steps** 1. Generate a short script with a visible learning purpose. 2. Read it aloud at a natural teaching pace and time the reading. 3. Mark awkward, jargon-heavy, or unsupported sentences; ask for one specific revision. 4. Keep a clean transcript and check captions if a recording is made. **Review** - Does the spoken version fit the intended time? - Would a novice understand the example? - Are causal statements appropriately qualified? - Does the transcript match the final spoken words? **Expected output:** drafts/association-script.md with visual cues, plus a clean transcript ready to accompany a recording **Sources** - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 33. Build a rubric from evidence **ASSESS AND FACILITATE · About 8 minutes** Start with the outcome and student task. Draft distinct criteria and observable descriptions of quality. **Make:** An analytic rubric **Check:** Can students tell levels apart? **Tool:** Cursor or Antigravity **Inputs:** Starter-kit COURSE_BRIEF.md and APPROVED_READING.md for the campus transport proposal exercise; the instructor's assignment requirements. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Build a draft analytic rubric for the campus transport proposal: students compare two options, recommend one, use at least two relevant pieces of supplied evidence, and acknowledge a tradeoff and a limitation. Preserve the actual course outcome and task constraints from the brief. Propose four distinct criteria: use of evidence and its limits, comparison of options, reasoning about tradeoffs, and clarity for the intended decision-maker. Draft four performance levels with observable descriptions specific to this task. Avoid labels such as 'excellent' as the only distinction and avoid counting citations as a substitute for judging their relevance. Do not assess grammar as a separate criterion unless it prevents meaning. Produce drafts/recommendation-rubric.md and a CSV version. Explain how each criterion maps to the outcome; leave weighting as an explicit instructor decision. Flag any criterion that goes beyond the supplied assignment. ``` **Steps** 1. Give the agent the actual assignment, outcome, and context. 2. Inspect one criterion across all performance levels for meaningful distinctions. 3. Check that different criteria are not scoring the same quality twice. 4. Save Markdown and CSV versions, then calibrate before using it. **Review** - Is every criterion tied to intended learning? - Are performance levels observable and distinguishable? - Is the same weakness counted twice? - Can a student use the descriptors to improve? - The descriptors explicitly preserve the required evidence, tradeoff, and limitation. **Expected output:** drafts/recommendation-rubric.md, drafts/recommendation-rubric.csv, and an outcome-to-criterion map **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart ## 34. Calibrate the rubric **ASSESS AND FACILITATE · About 9 minutes** Apply the draft rubric to two fictional samples. Compare ratings, explain disagreements, and revise ambiguous descriptors. **Make:** A calibrated draft rubric **Check:** Would two reviewers agree why? **Tool:** Cursor or Antigravity plus independent instructor/peer review **Inputs:** The draft rubric from activity 33 and starter-kit sample-responses.md, which contains fictional practice responses. Include APPROVED_READING.md and sample-data.csv to verify factual and numerical claims. ```text Read COURSE_BRIEF.md, APPROVED_READING.md, sample-data.csv, sample-responses.md, and drafts/recommendation-rubric.md. Use the supplied fictional practice responses for rubric calibration; preserve the source file and its synthetic status. First verify the sample claims and calculations against the supplied reading and data. In separate rating passes, apply the rubric criterion by criterion, quoting short evidence from each sample and noting uncertainty. Do not hide disagreement or average it away. If the supplied responses do not fit the current task, flag the mismatch before rating. Create drafts/rubric-calibration.md with the evidence for each rating, disputed boundaries, and proposed descriptor revisions. Preserve the original rubric; save the proposed revision as drafts/recommendation-rubric-v2.md. The instructor will independently rate the samples before accepting changes. Do not infer facts about real students or assign official grades. ``` **Steps** 1. Have the instructor or a colleague rate the supplied synthetic samples before seeing the agent's ratings. 2. Compare evidence and rationale by criterion, not only totals. 3. Identify one ambiguous descriptor and revise its boundary. 4. Re-rate the same samples with the revision and record whether the ambiguity improved. **Review** - Are sample responses explicitly synthetic? - Is each rating supported by observable evidence? - Do disagreements reveal ambiguous language? - Does the revision preserve the intended standard? **Expected output:** Criterion-level ratings of the supplied synthetic responses, a calibration log, and a revised rubric version **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart ## 35. Make a question bank **ASSESS AND FACILITATE · About 8 minutes** Generate varied questions aligned to one outcome. Include model answers, explanations, and the evidence each item tests. **Make:** A reviewed question bank **Check:** Does every answer hold up? **Tool:** Cursor or Antigravity **Inputs:** An approved teaching explanation and one measurable learning outcome. ```text Create a draft bank of eight questions for this outcome: students can distinguish an observed association from a warranted causal claim and identify a plausible alternative explanation. Use the instructor-approved explanation I supply; ask for it if absent. Include two short responses, two data-interpretation questions using clearly labelled synthetic data, two application scenarios, and two explanation questions. For each item provide the outcome, required cognitive process, correct or defensible model answer, rationale, acceptable alternatives, and the evidence a student's answer should reveal. Do not invent source citations. Put student-facing questions in drafts/question-bank.md and answers in drafts/question-bank-key.md. Add a CSV with stable item IDs. Independently check each answer and flag anything requiring instructor confirmation before release. ``` **Steps** 1. Supply a short approved source and one outcome. 2. Generate a varied bank and keep the key separate from the student version. 3. Solve several items yourself without looking at the answer key; inspect every key before release. 4. Revise ambiguity, accidental clues, or tasks that test a different outcome. **Review** - Does each item assess the stated outcome? - Are synthetic data identified? - Does every key survive independent solving? - Are alternative defensible responses acknowledged? - Does the student file omit answers? **Expected output:** drafts/question-bank.md, drafts/question-bank-key.md, and drafts/question-bank.csv **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart ## 36. Use misconceptions well **ASSESS AND FACILITATE · About 6 minutes** Write plausible distractors from common misunderstandings. Explain why each option is right or wrong without relying on tricks. **Make:** A diagnostic multiple-choice item **Check:** What would each choice reveal? **Tool:** Cursor, Antigravity, or a text assistant **Inputs:** A common misunderstanding from the instructor's teaching experience; no student identities needed. ```text Write one diagnostic multiple-choice question for first-year statistics. Misunderstanding to probe: a strong association alone proves that one measured variable causes the other. Use a short fictional campus example with a plausible common cause, and ensure the stem provides enough information to choose the best answer. Include four parallel options with one best-supported answer. Make each distractor reflect a different plausible reasoning error rather than a joke or a trick. For every option, explain the reasoning it may indicate and write helpful feedback. Do not claim that selecting an option establishes a student's belief with certainty. Save drafts/diagnostic-item.md and a separate instructor key. Add one follow-up 'why?' question that can distinguish guessing from reasoning. ``` **Steps** 1. Describe the reasoning error rather than merely requesting 'harder distractors'. 2. Read the stem without options to see whether the problem is clear. 3. Check that options are parallel and no accidental wording clue gives away the answer. 4. Use a justification or peer discussion to learn what the selection actually means. **Review** - Is there one best answer under the stated conditions? - Is every distractor plausible without being misleading? - Are response patterns interpreted cautiously? - Does feedback explain the concept instead of only naming right or wrong? **Expected output:** A diagnostic item, four option-level rationales, feedback, and a follow-up reasoning question **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf ## 37. Create a case with a decision **ASSESS AND FACILITATE · About 8 minutes** Build a fictional case with competing priorities, enough evidence, and a decision students must justify. **Make:** A discussion case **Check:** Is there room for reasoned disagreement? **Tool:** Cursor or Antigravity **Inputs:** A decision students should be able to justify. Self-contained fictional campus project values appear in the prompt. ```text Create a fictional campus sustainability case for a 20-minute first-year seminar. Students must recommend one project within a $60,000 budget. Use these invented options consistently: building lighting upgrade costs $50,000 and is estimated to avoid 18 tonnes CO2e/year; a sheltered cycle hub costs $60,000 with an uncertain estimate of 8–20 tonnes/year and access benefits concentrated among cyclists; a refill-station programme costs $35,000 with an uncertain estimate of 3–7 tonnes/year and convenient locations still to be decided. These are invented teaching assumptions, not real engineering estimates. Include a short scenario, a clear data table, stakeholder perspectives that avoid stereotypes, uncertainty, and three discussion prompts. Require a recommendation, a tradeoff, and one additional piece of evidence students would seek. Save drafts/campus-case.md and a separate facilitator guide. Do not force one predetermined correct answer. ``` **Steps** 1. Choose a decision that requires the target reasoning, not only recalling a fact. 2. Inspect all invented values for internal consistency. 3. Try to justify at least two different recommendations using the stated criteria. 4. Separate the student case from the facilitator's reasoning notes. **Review** - Is the decision explicit and feasible? - Are fictional assumptions visibly identified? - Can more than one option be defended? - Do stakeholder perspectives avoid caricature? - Is the required evidence of learning clear? **Expected output:** drafts/campus-case.md with a data table and decision task, plus drafts/campus-case-facilitator.md **Sources** - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ ## 38. Stage a useful disagreement **ASSESS AND FACILITATE · About 8 minutes** Design a structured debate with evidence requirements, assigned perspectives, and a reflection that allows views to change. **Make:** A classroom debate kit **Check:** Are students reasoning or performing? **Tool:** Cursor or Antigravity **Inputs:** The fictional campus case from activity 37, or another evidence-based decision case. ```text Turn drafts/campus-case.md into a structured classroom disagreement for groups of four. Preserve its fictional data and uncertainties. Assign analytical perspectives rather than personal identities: minimise emissions, broaden access, manage financial uncertainty, and test evidence quality. Each participant must cite two pieces of case evidence, state one limitation, and accurately summarise another perspective before responding. Create a 15-minute sequence with preparation, exchange, synthesis, and individual reflection; include a quiet written participation route. End with a recommendation that may differ from the assigned perspective and a sentence explaining what changed or why it did not. Save drafts/debate-kit.md with role cards, timings, student instructions, and instructor observation prompts. Avoid framing the exercise as winning at all costs. ``` **Steps** 1. Give every perspective access to the same evidence. 2. Require participants to restate another view accurately before rebuttal. 3. Leave time to change a recommendation after discussion. 4. Use a short individual reflection to capture learning beyond verbal confidence. **Review** - Does each perspective have legitimate evidence to consider? - Can quieter or remote learners contribute meaningfully? - Does the structure reward listening and revision? - Does the reflection reveal individual reasoning? **Expected output:** drafts/debate-kit.md with perspective cards, a timed sequence, evidence requirements, and a reflection prompt **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf ## 39. Rehearse a difficult conversation **ASSESS AND FACILITATE · About 8 minutes** Create role cards, a realistic opening, and a debrief for a professional conversation students need to practice. **Make:** A role-play activity **Check:** What will the debrief surface? **Tool:** Cursor or Antigravity; participants practise in pairs or triads **Inputs:** A professional communication outcome. Example: clarify a missed team deadline without making assumptions about a colleague. ```text Design a role-play for an undergraduate professional-communication course. Outcome: students can describe an observable problem, ask an open question, listen, and agree a concrete next step. Scenario: a project team has missed an internal deadline; one member assumes lack of effort, while the other has been working from an outdated brief. Create two role cards with limited information, a neutral opening line, an observer checklist tied to the outcome, and a five-minute debrief. Do not prescribe a single perfect script or require personal disclosure. Include an alternative where a learner annotates and revises a written dialogue to practise the same reasoning when appropriate. Save drafts/deadline-roleplay.md. Label the scenario fictional and keep the facilitator's hidden information separate from each role card. ``` **Steps** 1. Read the role cards separately so the information gap remains real. 2. Practise a two-minute opening while an observer records specific behaviours. 3. Pause and identify an assumption or missed question. 4. Repeat the opening with one change and debrief what became clearer. **Review** - Do role cards create a realistic information gap? - Is the observer looking for behaviours rather than personality traits? - Is there a concrete agreed next step? - Does the debrief connect the experience to the learning outcome? **Expected output:** Separate fictional role cards, an opening prompt, observer checklist, replay instruction, and debrief questions **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ - Krathwohl (2002), A Revision of Bloom's Taxonomy: An Overview: https://ou.edu/content/dam/assessment/docs/Theory%20Into%20Practice.pdf ## 40. Make feedback actionable **ASSESS AND FACILITATE · About 6 minutes** Use a fictional response and the rubric to draft feedback with one strength, one priority, and one next step. **Make:** A feedback model **Check:** Could the student act on it? **Tool:** Cursor, Antigravity, or a text assistant with the selected files supplied **Inputs:** COURSE_BRIEF.md, sample-responses.md, APPROVED_READING.md, and the rubric created earlier; use a provisional rubric only if that activity was skipped. Include sample-data.csv when checking numerical claims. ```text Read COURSE_BRIEF.md, APPROVED_READING.md, sample-data.csv, sample-responses.md and my rubric. Verify its factual and numerical claims against the reading and data. Select one fictional response. Draft no more than 120 words of feedback: one evidenced strength, one priority tied to a specific rubric criterion, and one concrete next step the learner can carry out. Identify the passage in the response that supports your feedback. Do not invent student intentions or refer to evidence the response does not contain. Do not assign a grade. Then propose a short revision task the learner could complete in five minutes. Save feedback-model.md. If no rubric exists, propose a small provisional set of criteria for my review before generating feedback. ``` **Steps** 1. Select a fictional response and one relevant rubric criterion. 2. Generate the short feedback model and its cited passage. 3. Compare each assertion with the actual response and reading. 4. Ask a colleague to carry out only the next-step instruction; observe whether it is clear. 5. Revise an overly broad instruction into a small action, then save the accepted example. **Review** - The claimed strength and problem can both be located in the fictional response. - The priority refers to the actual rubric and learning goal. - A learner can identify exactly what to revise. - The feedback does not introduce a new assessment criterion or unsupported judgement. **Expected output:** A short feedback model, linked to visible evidence, plus one achievable revision task. **Sources** - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/ ## 41. Build a one-page module **BUILD SOMETHING USABLE · About 12 minutes** Ask Cursor or Antigravity to turn your brief into a local web page with outcomes, explanation, practice, and an exit ticket. **Make:** A working HTML module **Check:** Can a student follow it alone? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** COURSE_BRIEF.md and APPROVED_READING.md; a new folder for this build. ```text Read COURSE_BRIEF.md and APPROVED_READING.md. Plan a 10-minute introductory learning module about evaluating a campus sustainability decision. Show the learning outcome, student journey, proposed sections, and acceptance checks before building. After I approve the plan, create module.html with an outcome, a concise explanation grounded only in the supplied reading, one worked comparison, a practice decision with explanatory feedback, and a two-sentence exit-ticket prompt. Label the authored teaching reading accurately; do not invent external citations. Include a source note and a printable view. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. Keep student answers only in page memory until reset or reload. Do not promise LMS gradebook integration. ``` **Steps** 1. Use Cursor Plan Mode, or Antigravity Planning with Request Review, for the proposed module. 2. Check whether the practice and exit ticket really demonstrate the chosen outcome. 3. Approve the plan and ask the agent to build module.html. 4. Double-click the file in a browser and complete the activity without instructor explanation. 5. Request one focused revision, such as clearer comparison criteria or better practice feedback. **Review** - The file opens offline without a terminal or account. - Every activity is traceable to the brief and supplied reading. - Practice feedback explains why a choice is defensible or incomplete. - Keyboard users can reach every control and see focus. - A new learner can complete the module and exit ticket unaided. **Expected output:** A local HTML module with explanation, meaningful practice, feedback, exit ticket, sources and print view. **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode - Google Antigravity Getting Started: https://www.antigravity.google/docs/getting-started - Google Antigravity Artifact Review: https://www.antigravity.google/docs/artifact-review - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 42. Make a concept interactive **BUILD SOMETHING USABLE · About 12 minutes** Build a small explainer with a slider and a prediction question. Keep the underlying relationship visible and testable. **Make:** An interactive concept demo **Check:** Does the interaction teach the concept? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** The workshop brief and the exact small population specified in the prompt; no real student dataset. ```text Plan and build sampling-demo.html, a small concept demonstration for how sample size affects the variability of sample means. Use the synthetic population [10,20,30,40,50] with equal probabilities and independent sampling with replacement. Let a labelled slider choose n from 1 to 25. Before displaying the results, ask the learner to predict whether the distribution of sample means will become wider, narrower, or stay similar as n increases. Keep that prediction visible alongside explanatory feedback. Generate 5,000 sample means using a documented deterministic seed for repeatable demonstration. Show a histogram, accessible data summary, and the reference relationship: population mean 30; population variance 200; standard deviation of the sample mean = sqrt(200/n). State that the simulation will vary around the theoretical result and that larger samples do not correct a biased sampling method. Keep the horizontal scale fixed between settings so comparisons are meaningful. Provide a reset button and a short explanation of why individual observations and sample means are different. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. Show the plan and three calculation checks before building. ``` **Steps** 1. Ask learners to predict what changes when n grows. 2. Inspect the plan for a fixed population, sampling with replacement, and a fixed chart scale. 3. Build and open the demo locally. 4. Compare n=1 and n=25; check the theoretical standard deviations against the displayed formula. 5. Have a colleague explain why the centre stays at 30 while the theoretical spread shrinks, and why a larger biased sample remains a problem. **Review** - The population mean is 30 and population variance is 200. - The theoretical spread is sqrt(200) at n=1 and sqrt(8) at n=25. - The histogram uses sample means rather than raw observations. - Axes remain comparable across slider settings. - The page explains sampling variability and does not equate larger n with removal of bias. - The slider is keyboard operable and its current value has a text label. **Expected output:** An offline interactive sample-mean demonstration with prediction, simulation, reference calculations and text alternative. **Sources** - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 43. Build a branching scenario **BUILD SOMETHING USABLE · About 14 minutes** Turn a case into three decision points with consequences and feedback. Map every path before generating the page. **Make:** A branching learning activity **Check:** Can every path reach a useful debrief? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** COURSE_BRIEF.md and APPROVED_READING.md; campus sustainability decision as the shared case. ```text Read the course brief and approved teaching reading. Design a campus sustainability decision scenario with exactly three sequential decision points and two options at each point. Keep all eight choice sequences reachable. Branch-specific consequences may differ, but every sequence must reach a debrief about evidence, stakeholder priorities and trade-offs. First save scenario-map.md with each decision, the two options, immediate feedback, possible accumulated consequences and all eight terminal sequences. Use fictional context and no unsupported claims about a real institution. After I approve the map, build scenario.html. Include a visible record of the learner's choices, a debrief, and restart. Ask the learner to defend one trade-off; do not reduce a defensible policy choice to a single simplistic right answer. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. Also save a path-test.md table recording the eight paths and whether each reaches its debrief. ``` **Steps** 1. Review the map before generating the interactive page. 2. Check that alternatives are plausible and their consequences are consistent with the reading. 3. Build the page, then deliberately choose the least expected path. 4. Test all eight sequences and compare them with the map. 5. Revise one shallow feedback message to explain a trade-off and invite justification. **Review** - All eight choice sequences terminate without a dead end. - Consequences do not contradict earlier choices. - The final debrief refers to the learner's actual decisions. - Restart clears the old state. - Feedback uses the supplied evidence and makes room for justified alternatives. **Expected output:** A mapped and tested branching scenario, with feedback on choices and a useful debrief on every path. **Sources** - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 44. Let students explore data **BUILD SOMETHING USABLE · About 12 minutes** Use a small synthetic dataset to build a filterable chart and interpretation questions. Check the chart against the source rows. **Make:** A classroom data explorer **Check:** What could the display mislead about? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** sample-data.csv, the provided synthetic campus transit dataset, and COURSE_BRIEF.md. ```text Read sample-data.csv first and report its actual columns, units, row count and missing values. Propose one learning question this synthetic campus transit dataset can support. Do not invent columns or treat the observations as causal evidence. After I approve, build transit-explorer.html: embed the exact CSV rows in the file; offer one meaningful category filter; show a clearly labelled bar chart, count of included rows, the displayed summary calculation, and the matching source rows in an accessible table. Use a zero baseline for bars. Distinguish totals, averages and percentages and state the denominator. Include three interpretation questions and a revealable teaching note explaining one limitation. Handle an empty result with a readable message. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. If the file lacks the fields needed for the requested chart, explain the mismatch and propose a supported chart instead. Save data-check.md comparing three displayed results with calculations from the underlying rows. ``` **Steps** 1. Inspect the actual data schema before specifying a chart. 2. Choose a single classroom question and meaningful filter. 3. Build the offline page and inspect a filtered view. 4. Recalculate three displayed values from source rows. 5. Ask learners to identify a conclusion the data cannot support. **Review** - Every plotted value matches the embedded source data. - Labels identify units, aggregation and denominator. - The source table updates with the filter. - The chart makes clear that the dataset is synthetic. - An empty filter state is usable. - The discussion avoids causal conclusions from descriptive observations. **Expected output:** A filterable local chart with source rows, explicit calculations and interpretation questions. **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 45. Build practice with feedback **BUILD SOMETHING USABLE · About 10 minutes** Create a local practice quiz with explanations and a reset button. Test correct, incorrect, and unanswered responses. **Make:** A self-check activity **Check:** Does the feedback explain the reasoning? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** APPROVED_READING.md and COURSE_BRIEF.md; instructor-reviewed question ideas from the assessment activities. ```text Read the brief and approved teaching reading. Draft four formative practice questions on comparing evidence and trade-offs in a campus sustainability decision. For each, give the intended outcome, correct or best-supported answer, explanation, and the misconception behind each distractor. Make one question ask students to interpret a small piece of supplied evidence rather than recall terminology. Show me the question table before building. After review, create practice.html with one question at a time or clearly separated fieldsets, labelled answer controls, a Check answer action, explanatory feedback for every option, and Reset. Handle an unanswered question by asking the learner to make a choice without marking it wrong. Let learners try again and read the reasoning. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. Do not collect names, save grades, or present this as a secure examination system. Save tests.md with checks for correct, incorrect, unanswered, changed-answer and reset states. ``` **Steps** 1. Review the four questions and distractor explanations for accuracy. 2. Build the local practice page. 3. Try each answer option, including every distractor. 4. Check an unanswered question, change an answer after feedback, and reset. 5. Ask whether the explanation helps the student solve a new problem rather than merely revealing a letter. **Review** - Each answer key is checked against the approved material. - Every distractor has accurate and useful feedback. - Unanswered items are handled explicitly. - Changing an answer does not leave stale feedback. - Reset clears selections and messages. - Keyboard and screen-reader labels identify the controls. **Expected output:** An offline self-check activity with reviewed questions, explanatory feedback and dependable reset behaviour. **Sources** - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 46. Make a classroom activity board **BUILD SOMETHING USABLE · About 10 minutes** Build a timer, group instructions, and a sequence of prompts for one class activity. Test it on the projector. **Make:** A facilitator display **Check:** Can the room use it without extra explanation? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** COURSE_BRIEF.md, APPROVED_READING.md and a planned small-group comparison activity. ```text Build activity-board.html for a classroom comparison of two campus sustainability options. Propose three stages: individual prediction, small-group evidence comparison, and whole-class debrief. Each stage needs one short instruction, a visible required output, and a suggested duration that I can change. Include large readable controls for Start, Pause, Reset, Next and Previous. Calculate elapsed time using a real time reference rather than subtracting one second per interval; never show a negative countdown. When time ends, keep the current instructions visible until I advance. Add an untimed mode and a reduced-motion presentation. Do not depend on sound. Include a print view with the full instructions and prompts. Use one self-contained HTML file with inline CSS and JavaScript. It must open by double-clicking the file, with no build step, external library, network request, API key, account, or learner-data collection. Use semantic headings, explicit form labels, keyboard controls, visible focus, readable contrast, and feedback that does not depend on colour alone. First show the three-stage activity plan for review. Then test pause/resume, reset, moving back, and completion. ``` **Steps** 1. Review the task and the required student output at every stage. 2. Build and open the board locally. 3. Switch to the actual projector or a comparable large-screen view. 4. Pause during the timer, resume, reset and move between stages. 5. Ask a participant at the back of the room to explain the current task from the display alone. **Review** - The room can read the main instruction and required output. - Pause, resume, reset and navigation behave consistently. - The timer does not auto-dismiss instructions. - The activity also works in untimed mode. - Controls are labelled and keyboard accessible. - The final stage elicits a reasoned recommendation. **Expected output:** An offline facilitator board with stage instructions, visible outputs, flexible timing and print version. **Sources** - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ ## 47. Try the student journey **REVIEW AND RELEASE · About 8 minutes** Ask a reviewer to follow the resource as a new learner. Then walk through it yourself and record where you get stuck. **Make:** A usability issue list **Check:** What did the agent overlook? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** One generated module, scenario or practice page; the intended learning goal; an empty review log. ```text Review this local learning resource as a first-time learner using only the instructions visible in the resource. Do not edit files yet. Begin at the opening screen and follow the essential task through to its final learning output. If browser tools are available, actually operate the controls and distinguish observed behaviour from code inspection. Try one incorrect or unexpected action and one restart. Save student-journey-review.md with: step attempted, expected behaviour, observed behaviour, evidence or location, impact on learning, and suggested repair. Mark anything you could not test as unverified. Then separately compare the completed learning output with the intended outcome. Do not declare success solely because the page loads. ``` **Steps** 1. Give the reviewer the resource and learning goal, not an extra verbal walkthrough. 2. Read the issue list and separate observed failures from guesses. 3. Complete the same route yourself using the visible instructions. 4. Add one issue that the agent missed, or confirm none was found. 5. Choose the two most consequential repairs for the next pass. **Review** - The test reaches the intended student output, not only the homepage. - Observed behaviour and untested assumptions are labelled separately. - At least one unexpected path is checked. - Issues describe a learning impact and a concrete repair. - A human has walked the core route. **Expected output:** A prioritized usability issue list supported by specific steps and observations. **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Google Antigravity IDE Browser Overview: https://www.antigravity.google/docs/ide/browser - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ ## 48. Check access in practice **REVIEW AND RELEASE · About 10 minutes** Test keyboard use, headings, contrast, text alternatives, and zoom. Use automated checks to support your own inspection. **Make:** An accessibility repair list **Check:** Can learners complete the essential task? **Tool:** Cursor or Antigravity with the workshop folder open; any modern browser **Inputs:** The local HTML resource plus a browser; an optional accessibility checker if already available. ```text Inspect this learning resource for access barriers. First audit the HTML for heading structure, form labels, text alternatives, table headers, language, and meaningful reading order. Then, where browser tools permit, test the entire essential task by keyboard, confirm visible focus and no keyboard trap, and inspect the layout at 200 percent zoom and a narrow viewport. Check text contrast against WCAG 2.2 AA thresholds: 4.5:1 for normal text and 3:1 for qualifying large text. Check that feedback does not rely only on colour. Complex charts need a useful text explanation or data table. Use any available automated checks as supporting evidence; do not claim that an automated pass proves complete conformance. Save accessibility-review.md with observed barrier, affected task, proposed fix, verification method and status. Make no changes until I select the repair list. After repair, retest the essential task. ``` **Steps** 1. Put the mouse aside and complete the resource using the keyboard. 2. Zoom to 200 percent and inspect the reading order and controls. 3. Review image purpose, alternative text, chart summaries and tables. 4. Compare manual findings with any available automated report. 5. Repair the highest-impact barrier and repeat that exact task. **Review** - Every essential control works by keyboard. - Focus is visible and moves in a meaningful order. - Headings, labels and data tables convey structure. - Text remains usable at zoom and a narrow viewport. - Important graphics have an appropriate equivalent. - Claims distinguish checks performed from items not tested. **Expected output:** An evidence-based accessibility repair list and a retested essential task. **Sources** - Web Content Accessibility Guidelines 2.2: https://www.w3.org/TR/WCAG22/ - W3C WAI Images Tutorial: https://www.w3.org/WAI/tutorials/images/ - W3C WAI Complex Images: https://www.w3.org/WAI/tutorials/images/complex/ - W3C WAI Making Events Accessible: https://www.w3.org/WAI/teach-advocate/accessible-presentations/ ## 49. Revise, verify, package **REVIEW AND RELEASE · About 8 minutes** Fix the important issues, rerun the task, and package the approved files with sources, instructions, and a change log. **Make:** A teaching-ready release **Check:** Would you use this with your class? **Tool:** Cursor or Antigravity; browser; local file manager **Inputs:** Draft resource, student-journey-review.md, accessibility-review.md, approved reading, brief and instructor-selected repair list. ```text Read the brief, current resource and review logs. Apply only the repairs I have selected and preserve the accepted content. Work on a new release candidate so the prior version remains available. After each consequential repair, rerun the affected student task and record the actual result. Do not silently mark untested checks as passed. Prepare a local release/ folder containing the finished resource, required local assets, README.md with opening and classroom-use instructions, SOURCES.md with accurate source and asset provenance, CHANGELOG.md, and REVIEW.md with tests performed and remaining limits. Check for broken relative links and unexpected network dependencies. Reopen the packaged copy from release/ to confirm it works independently of the draft folder. Do not upload or publish it. End with a concise review summary for my teaching decision. ``` **Steps** 1. Select repairs based on learning impact and access barriers. 2. Create a release candidate while retaining the previous draft. 3. Retest repaired behaviour and the complete core activity. 4. Open the packaged resource from its release folder. 5. Make the instructor's final decision: use, revise again or set aside. **Review** - The packaged resource opens from its own folder. - All required local assets and relative links work. - Selected repairs are verified with actual tasks. - The sources and provenance describe the materials honestly. - Known limits are retained in the review record. - No automatic publishing or student-data collection was added. **Expected output:** A self-contained release folder with usable teaching files, instructions, sources, change log and review record. **Sources** - Cursor Quickstart: https://cursor.com/docs/get-started/quickstart - Google Antigravity Artifacts: https://www.antigravity.google/docs/artifacts/ ## 50. Run your own production sprint **YOUR NEXT BUILD · About 20 minutes** Choose one real need. Brief it, plan it, build it, critique it, revise it, and make the final teaching decision. **Make:** One reviewed course resource **Check:** What will you build this week? **Tool:** Cursor, Antigravity or another approved agent workspace; browser for final review **Inputs:** One real teaching need, the instructor's approved materials, and the workshop brief as a reusable template. ```text Help me run a small course-resource production sprint. My teaching need is [need]. The audience is [learners], the learning outcome is [outcome], available time is [time], and approved inputs are [files]. The final resource should be [one deliverable]. First read the inputs and restate the task, evidence of learning and success criteria. Propose a bounded plan and wait for my review. After approval, build the smallest complete version. Use separate writer and critic passes, or separate subagents if available; give both the same brief. The critic should identify specific evidence, alignment, usability and access issues. Revise the important findings, test the core learner task, and record what remains unverified. For an interactive resource, use self-contained local HTML/CSS/JavaScript with no build tools or external services. Package the result with sources and opening instructions. Keep the final decision about teaching use with me. ``` **Steps** 1. Choose a need that fits one concrete resource, not an entire course redesign. 2. Write the audience, outcome, inputs, time and success criteria. 3. Review the plan, then build a minimum complete version. 4. Run a focused critique and request one meaningful revision. 5. Test the learner task and decide whether to use the result. 6. Name the next resource or update you will attempt this week. **Review** - The need and deliverable are small enough to finish. - The outcome describes learner performance. - The critique uses explicit criteria and concrete evidence. - A revision changes something that matters for learning. - The core learner task has been tested. - The instructor records a final teaching decision. **Expected output:** One reviewed course resource and a repeatable process for the next build. **Sources** - Cursor Plan Mode: https://cursor.com/help/ai-features/plan-mode - Cursor Subagents: https://cursor.com/docs/subagents - Google Antigravity Artifact Review: https://www.antigravity.google/docs/artifact-review - CAST UDL Guidelines 3.0: https://udlguidelines.cast.org/