assessment-design

assessment-design is a skill for Claude Code, Codex from scibly-dev/skills. It costs 112 tokens per session (1,089 once invoked), scanned A, original, MIT.

A guide for using AI to create learning assessments, including knowledge checks, scenario questions, multiple-choice items, distractors, and teaching feedback. It uses Bloom's taxonomy, a framework for choosing how deeply learners must understand or apply a topic.

In plain words
What is it for?
Use it to design quizzes, tests, application-level questions, scenario-based items, and explanations for learners. It is aimed at instructional designers and course developers.
Why use it?
It addresses the common problem of AI-generated questions testing simple memorisation instead of practical understanding. The process helps match questions to the learner's context and the intended level of thinking.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/scibly-dev/skills/assessment-design
Any agent
npx skills add scibly-dev/skills --skill assessment-design
Clone the repo
git clone --depth 1 https://github.com/scibly-dev/skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for assessment-design

README.md
[![agentmods](https://agentmods.dev/badge/skills/scibly-dev/skills/assessment-design.svg)](https://agentmods.dev/skills/scibly-dev/skills/assessment-design)
Your own site
<a href="https://agentmods.dev/skills/scibly-dev/skills/assessment-design"><img src="https://agentmods.dev/badge/skills/scibly-dev/skills/assessment-design.svg" alt="Measured on agentmods" height="20"></a>
Per session 112 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,089 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00112 $0.01089
Opus 5 $0.00056 $0.00544
Sonnet 5 $0.00022 $0.00218
Haiku 4.5 $0.00011 $0.00109

Measured 6d ago against content hash 63cdaa73f765, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

assessment-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/assessment-design/SKILL.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Assessment Design with AI

Most AI-generated questions are too easy — they test recognition, not learning. Worse, a common AI failure mode is testing the wrong thing entirely: quizzing an exact percentage, date, or phrase pulled from the source text instead of whether the learner can actually use the underlying skill. A learner can memorize "73%" and still fail to apply the concept it came from; that's not transfer, it's trivia. This skill shows you how to use Bloom's taxonomy as a prompt lever to get application-level questions with quality distractors and feedback that actually teaches.

Step 1 — Understand the assessment context

Ask the user:

  1. What's the topic? Be specific — not "leadership", but "giving constructive feedback to a peer who missed a deadline".
  2. What Bloom's level? If unsure:
    • Remember / Understand: learners recall or explain concepts → compliance or foundational knowledge
    • Apply: learners use knowledge in a new situation → skills and procedures
    • Analyze / Evaluate: learners break down situations or make judgments → complex decision-making
  3. Question format? MCQ, scenario-based (situation + question), true/false with explanation, short answer, matching, drag-to-sequence.
  4. How many questions? And for what purpose — formative check, end-of-course test, certification?
  5. What does the learner already know? Helps calibrate how tricky the distractors should be.

Step 2 — Build the assessment generation prompt

Construct this prompt:

ASSESSMENT GENERATION PROMPT

Role: You are an expert assessment designer and instructional design specialist.
Topic: [specific topic]
Target Bloom's level: [Remember / Understand / Apply / Analyze / Evaluate]
Learner: [role + experience level]
Question format: [MCQ / scenario-based / true-false / matching]
Number of questions: [N]

For each question:
- Write a stem that presents a realistic situation or judgment call — avoid "which of the following" where possible.
- [If MCQ] Write 4 answer options: 1 correct, 2 plausible distractors reflecting common mistakes, 1 tempting shortcut with a hidden flaw.
- [If scenario-based] Start with a 2-sentence situation before the question.
- Write feedback for each option: why it's correct or why it's a common mistake (2 sentences max per option).
- Tag each question with its Bloom's level.

Avoid: trivially obvious wrong answers, questions answerable by scanning the course text without thinking, trick questions, double negatives.

Read the full file on GitHub · 91 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 91 lines · 112 tokens per session scan A 63cdaa73f765

Subscribe to this mod's changes

assessment-design is a skill published in the GitHub repository scibly-dev/skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 112 tokens to every session and 1,089 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories