Legal-document classification
Privileged vs. discoverable. Material vs. immaterial. Confidential vs. public. The edge cases are where the training actually pays off.
Coming soon
Have Claude or ChatGPT generate the edge cases, learners drag and drop.
Build drag-and-drop categorization activities in plain language. The AI generates plausible edge cases your learners will actually disagree about, so the activity teaches discrimination, not pattern-matching.
What is it?
What is a sort-and-categorize activity?
A drag-and-drop training activity where learners place items into categories. The pedagogical value comes from the edge cases: items where the categorization is genuinely ambiguous and forces the learner to apply the underlying principle, not just spot the obvious matches.
-
Drag-and-drop interface with custom categories and items you describe in plain language -
AI generates edge-case items that genuinely require applying the rule, not just keyword-matching -
Per-category feedback explains why each item belongs where it does -
Replayable with different item mixes drawn from a larger pool -
Exports as a SCORM 1.2 package for Cornerstone, Moodle, Canvas, TalentLMS, and every LMS -
Same bundle runs standalone in any browser for self-study without an LMS
How it works
How to build a sort-and-categorize activity
in four steps.
From a plain-English description to a SCORM-ready drag-and-drop activity. No template editor, no item-bank spreadsheet, no manual edge-case authoring.
01
Describe the categories and the principle
Tell your coding agent what the categories are and what rule distinguishes them. Material vs. immaterial, capital vs. operating, urgent vs. important.
02
Let the AI generate edge cases
The tool generates a pool of items, including deliberately ambiguous ones that learners will get wrong if they pattern-match instead of applying the principle.
03
Refine the item pool in chat
"Make the financial items harder." "Add red-herring items that look like Category A but are actually Category B." Iterate by talking, not by spreadsheet editing.
04
Export as a SCORM 1.2 package
One command produces a SCORM 1.2 zip. Upload to any SCORM LMS. Scores post back to the gradebook automatically.
Use cases by role
Sort-and-categorize activities
for every taxonomy that matters.
The same toolchain powers legal classification training, accounting taxonomy practice, clinical triage drills, prioritization training, and security-incident classification.
Accounting taxonomy practice
Capital vs. operating expense. Revenue vs. deferred revenue. Accrual vs. cash. Real exam questions feature ambiguous items; your training should too.
Clinical triage prioritization
Sort presenting symptoms into urgency tiers. The edge cases mirror the real triage decisions that get clinicians into trouble.
Security-incident classification
Incident severity, data-classification labels, what-counts-as-a-breach. Sort-and-categorize sharpens the judgment calls that policy can't fully specify.
Eisenhower-matrix prioritization
Urgent / important / both / neither. Learners sort real workplace items and see where their instincts diverge from the principle.
Taxonomy training
Cell type, chemical class, programming-paradigm classification. The format works for any taxonomy worth memorizing through application rather than rote.
Principle-graded sorting
Every placement graded against the underlying principle, not a keyword match.
When the activity ends, the AI scores each placement against the rule you defined, weighted by item difficulty. Mistakes on obvious items count less than mistakes on edge cases, where the actual principle is tested.
Update categories, principles, or item difficulty any time. Re-export the SCORM package and the next attempt grades against the new rules.
AI-generated edge cases
The tool deliberately generates items that look like one category but belong to another, forcing learners to apply the underlying principle.
Per-item feedback
Wrong answers come with a one-line explanation of why the item belongs where it does. Right answers can include a confirming rationale too.
Randomized pools
Every attempt draws a different mix from a larger pool. Memorizing the right answers becomes a worse strategy than learning the principle.
Session complete
Your Debrief
9/10
★★★★★★★★★☆
Total 9 / 10
Correctly placed the two red-herring items that look like Category A but follow the Category B principle.
Applied the timing-vs-substance rule consistently across the harder financial items.
Recovered after the first miss by correctly placing the next three edge-case items.
Plain-English authoring
Author sort-and-categorize activities in plain English. No spreadsheet required.
Tweak categories, add edge cases, raise the difficulty, or rewrite the principle, just describe the change in your own words. Your coding agent rebuilds the item pool and re-exports the SCORM package without you ever editing a CSV.
When you're happy, export as a SCORM 1.2 package for your LMS, or use the same HTML bundle standalone.
Add five more edge-case items that look like operating expense but are actually capital.
Added 5 borderline items. Each surfaces the asset-life and dollar-threshold tests learners need to apply.
Make the easy items easier, current cohort gets 100% on them.
Lowered easy-tier difficulty. Top 8 items now have unmistakable signal words.
Export as SCORM.
✓ sort-and-categorize.zip ready (SCORM 1.2, 14 KB)
Categorization Library
Start from real classification
decisions.
Paste any prompt into your coding agent to get a complete sort-and-categorize activity as a single self-contained HTML file. Adjust categories, item pool, or feedback in plain language.
Accounting
Capital vs. operating expense
A 20-item drag-and-drop where learners sort real-looking expenses across the capital/operating boundary, with edge cases around asset life and dollar thresholds.
Prompt
Using /sort-and-categorize, build a 20-item activity sorting expenses into capital vs operating. Score on edge cases involving asset life and dollar thresholds.
Legal
Privileged vs. discoverable
An 18-item activity classifying documents as privileged, work-product, or discoverable. Edge cases probe waiver, dual-purpose communications, and joint-defense exceptions.
Prompt
Using /sort-and-categorize, build an 18-item privilege classification activity. Score on waiver, dual-purpose, and joint-defense edge cases.
Clinical
Triage urgency tiers
A 16-item activity placing presenting symptoms into ESI levels 1–5. Edge cases probe atypical presentations and 'sick-looking but low-acuity' patterns.
Prompt
Using /sort-and-categorize, build a 16-item ESI triage activity. Score on atypical presentations and overtriage avoidance.
Security
Data classification labels
A 15-item activity labeling sample documents as public, internal, confidential, or restricted. Edge cases probe customer PII, financial forecasts, and HR notes.
Prompt
Using /sort-and-categorize, build a 15-item data-classification activity. Score on customer PII, financial-forecast handling, and HR-note labels.
Management
Eisenhower-matrix prioritization
A 12-item activity placing real workplace tasks into urgent/important quadrants. Edge cases probe the difference between actually-urgent and feels-urgent.
Prompt
Using /sort-and-categorize, build a 12-item Eisenhower-matrix activity. Score on distinguishing actually-urgent from feels-urgent.
STEM
Programming paradigm classification
A 14-item activity sorting code snippets into OO, functional, procedural, and declarative paradigms. Edge cases probe multi-paradigm snippets.
Prompt
Using /sort-and-categorize, build a 14-item programming-paradigm activity. Score on OO vs functional vs procedural vs declarative, with multi-paradigm edge cases.
SCORM & LMS
SCORM 1.2 sort-and-categorize activities for any LMS.
Every sort-and-categorize activity exports as a standards-compliant SCORM 1.2 package. Upload the zip, assign it like any other course activity, and per-category scores flow back to the gradebook automatically.
-
Single SCORM 1.2 zip, upload to your LMS, no integration work, API keys, or custom JavaScript -
Completion, score, and time-on-task post back via cmi.core.score.rawandcmi.core.lesson_status -
Tested with Cornerstone OnDemand, Moodle, Canvas LMS, TalentLMS, Docebo, Brightspace, Absorb, 360Learning, SuccessFactors, and Workday Learning -
Same self-contained HTML bundle runs standalone on a public page, an intranet, or as an embedded iframe -
Accessibility-first drag-and-drop: keyboard navigation, screen-reader labels, and high-contrast mode included by default -
Grading runs inside the bundle, no middleware, no data warehouse, no analytics SDK required
FAQ
Frequently asked questions about
sort-and-categorize activities.
Authoring, edge cases, LMS compatibility, accessibility, model choice, answered directly.
A sort-and-categorize activity is a drag-and-drop training exercise where learners place items into categories. The pedagogical value comes from edge cases, items where the categorization is ambiguous and forces the learner to apply the underlying principle rather than pattern-matching.
Multiple-choice quizzes test recognition. Sort-and-categorize activities test judgment under ambiguity, especially when the item pool is full of edge cases. They're a better fit for taxonomies where the principle matters more than the specific labels.
Yes. The activity supports any number of categories. Common patterns are two-bucket (yes/no, valid/invalid), three-tier (urgent/important/neither), and four-quadrant matrices (Eisenhower, RICE, BCG).
Yes. The drag-and-drop UI supports touch, mouse, and keyboard input. Layout adapts to phone-sized screens.
Yes. Every activity exports as a SCORM 1.2 package, which works in Cornerstone, Moodle, Canvas, TalentLMS, Docebo, Brightspace, and every other SCORM-compliant LMS. Scores post back automatically.
Each item placement is scored against the correct category. The total score is the percentage of items placed correctly, with per-category breakdown so instructors see which distinction the learner struggles with. Edge cases can carry higher weight.
Given your categories and the underlying principle, the AI generates items that surface real ambiguity, things that look like one category but belong to another, things that depend on a clarifying detail. You review and approve before publishing.
Sort & Categorize is in active development. It will install via your coding agent with a single command, the same way the existing role-play skill works today. Add your email below and we'll let you know the day it ships.