Exam and assessment designer
Skill alainlebret/claude-agents/higher-ed-teaching-agents/skills/exam-and-assessment-designer
Modular skill and workflow ecosystems for AI coding assistants. Structured, multi-agent pipelines for specialized professional domains.
npx -y skills add alainlebret/claude-agents --skill exam-and-assessment-designerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Designs rigorous assessments for higher education and professional reskilling, including written exams, machine exams, quizzes, oral exams, take-home assignments, and retakes. Use when an instructor or trainer needs a scientifically sound, level-calibrated, feasible, and gradeable assessment with subject, instructions, rubric hooks, and correction guidance.
SKILL.md
2.3 KB, as published. Nobody here has run it
Exam and Assessment Designer
Goal
Create assessments that measure what was actually taught, remain feasible in the available time, and support fair grading.
Inputs
Read mission.json first. Then read any existing course, TD, TP, syllabus, prior exam, or institutional constraints.
Produce
Depending on the request, generate:
- assessment blueprint;
- subject statement;
- teacher version and student version when relevant;
- correction guidance;
- expected answers or reference solution outline;
- anti-ambiguity notes;
- grading hooks for
rubric-builder.
Assessment types supported
- written exam;
- machine exam;
- oral exam;
- quiz / QCM;
- take-home assignment;
- retake exam;
- diagnostic assessment;
- professional certification-style task.
Design rules
- Only assess notions that are explicitly taught or formally listed as prerequisites.
- Keep the wording precise, short, and unambiguous.
- Separate what is mandatory from what is optional.
- Calibrate the workload against the real exam duration.
- Distinguish knowledge recall, method application, reasoning, and technical execution.
- For machine exams, specify environment assumptions and recovery paths.
- For professional reskilling, prefer authentic tasks over purely academic tricks.
Mandatory checks
Before validating an assessment, verify:
- alignment with learning outcomes;
- level appropriateness;
- feasibility under time pressure;
- completeness of instructions;
- existence of a defensible grading path;
- consistency between subject, expected solution, and later rubric.
Anti-failure rules
- Refuse subjects with hidden dependencies.
- Refuse trick questions that reward guessing more than competence.
- Refuse questions requiring unstated notation or vocabulary.
- Flag topics assessed but not prepared by prior activities.
- Flag machine exams that depend on fragile tooling without fallback.