Aggregate cohort data into privacy-preserving evidence blocks for teaching decisions.
npx skills add https://github.com/YujxZJCN/teaching-skills --skill cohort-analyst
Please help me install this Agent Skill. Skill: cohort-analyst Source: https://github.com/YujxZJCN/teaching-skills/tree/main/cohort-analyst Command: npx skills add https://github.com/YujxZJCN/teaching-skills --skill cohort-analyst
This skill turns the data a professor already holds about a class into aggregated, privacy-preserving evidence that informs course design and weekly-lifecycle decisions, avoiding individual-student claims while enabling data-informed actions.
Design a concise ungraded diagnostic and run a cohort-profile analysis to produce a planning-ready, privacy-preserving evidence block.
High-intent search queries and answers about installing and using this skill.
To aggregate cohort data for teaching decisions, compute per-concept readiness distributions and identify two-tier misconception prevalence at the cohort level. This generates a passport-ready evidence block and actionable recommendations for course design and pacing without exposing individual student data.
Analyzing class diagnostic data without exposing individual student information requires aggregating findings into counts, proportions, and distribution sketches. This privacy-preserving approach outputs instrument-strength caveats and cohort-level evidence blocks while ensuring no individual-level data is ever written.
To identify cohort misconceptions for reteaching strategies, calculate the prevalence and heterogeneity shapes of two-tier misconceptions across the entire class. This aggregated diagnostic data illuminates where group-level understanding lies, allowing you to tailor grouping, pacing, and reteach needs accordingly.
Yes, you can generate a passport-ready evidence block from ungraded diagnostics by running a cohort-profile analysis. This produces a learner_profile.cohort_evidence block containing known difficulties, aggregated findings, and recommended actions that directly communicate evidence and caveats to course designers and instructors.
Cohort analytics applies to the design, build, and deliver phases for entire cohorts in higher education. It transforms existing class data into aggregated, privacy-preserving evidence that informs weekly-lifecycle decisions and course design actions throughout these phases without making individual student claims.
You should not use cohort-level aggregation when you need to make individual student claims or track single-learner progress. This approach is strictly designed to satisfy privacy requirements by outputting aggregated findings and evidence blocks while ensuring no individual-level data is written to any output.
ya-wayProfile raw 10x Genomics single-cell matrices to output QC distributions as JSON.
cless75Generate learner profiles and cohort insights from transcripts and self-assessments.
synaptiaiGround claims against evidence and output status, confidence, and gaps.
rcmaplesGuide privacy-conscious Fullstory data handling with GDPR, HIPAA, PCI, and CCPA alignment.
