Stats
Actions
Tags
From altistar
Evaluate student performance by analyzing course progress, quiz scores, and conversation transcripts. Generates evidence-based reports with pedagogical analysis and intervention recommendations for teachers.
How this skill is triggered — by the user, by Claude, or both
Slash command
/altistar:evaluate-studentThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill helps teachers evaluate student performance on the AltiStar platform. It analyzes course progress, quiz scores, activity patterns, and conversation transcripts to produce structured, evidence-based reports with actionable recommendations.
This skill helps teachers evaluate student performance on the AltiStar platform. It analyzes course progress, quiz scores, activity patterns, and conversation transcripts to produce structured, evidence-based reports with actionable recommendations.
When this skill is invoked, follow these steps:
Ask the user which type of report they need:
| Report Type | Purpose | Best For |
|---|---|---|
| Individual Student Report | Deep dive on one student's performance | Parent meetings, 1:1 check-ins, IEP reviews |
| Course Performance Report | Class-wide analysis of a specific course | Curriculum review, identifying struggling students |
| Intervention Alerts | Quick scan for students needing attention | Weekly check-ins, early warning detection |
Based on the report type, ask the user:
Follow the data collection flows below for each report type.
Apply the frameworks from pedagogy-guide.md to interpret the data:
Use the report templates below to produce the final output.
Use the altistar MCP server tools for data collection:
| Tool | Description | Use When |
|---|---|---|
list_users | List students/teachers with tag filtering | Finding students by name or tag |
get_user | Get full student profile (learning goals, challenges) | Individual student deep dive |
list_courses | List available courses | Course selection |
list_enrollments | List enrollments with filtering by course/student/teacher/status | Finding enrollments to analyze |
get_enrollment_progress | Per-resource scores, completion, attempts for an enrollment | Core performance data |
get_student_stats | 30-day averages, streak, all-time totals | Quick performance overview |
get_student_activity | Paginated activity log with scores and summaries | Recent session history |
get_transcript | Full conversation transcript with quiz responses | Deep engagement analysis |
1. list_users → identify the student (search by name or browse)
2. get_user(id) → full profile including learning goals, challenges, interests
3. list_enrollments(student=studentId) → all their enrollments
4. For each active enrollment:
get_enrollment_progress(enrollmentId) → per-resource scores and completion
5. get_student_stats(studentId) → 30-day overview, streak, totals
6. get_student_activity(studentId, limit=20) → recent sessions
7. For 2-3 key sessions (lowest scores or most recent):
get_transcript(progressId) → deep analysis of interaction quality
1. list_courses → identify the course
2. list_enrollments(course=courseId) → all enrolled students
3. For each enrollment:
get_enrollment_progress(enrollmentId) → per-resource scores
4. For flagged students (score <70% or high attempts):
get_student_activity(studentId, limit=5) → recent sessions
get_transcript(progressId) → selective deep dive (1-2 sessions max)
1. list_enrollments(status=active) → all active enrollments
(or filter by teacher if the user wants only their students)
2. For each enrollment:
get_enrollment_progress(enrollmentId) → check for warning signs
3. For flagged students:
get_student_stats(studentId) → confirm patterns
get_student_activity(studentId, limit=5) → recent history
Important token management: Never fetch all transcripts at once. Transcripts are large (JSONB). Only fetch them for 2-3 key sessions per student, selected based on the quantitative data.
# Student Performance Report: [Student Name]
**Date**: [Date]
**Report Period**: [Date range or "All time"]
## Student Profile
- **Learning Goals**: [From profile]
- **Learning Challenges**: [From profile]
- **Interests**: [From profile]
## Performance Summary
| Metric | Value | Assessment |
|--------|-------|------------|
| Mastery Rate (≥80%) | X/Y resources (Z%) | [Strong/Developing/Needs Support] |
| 30-Day Average Score | X% | [ZPD classification] |
| Current Streak | X days | [Engagement level] |
| Course Completion | X/Y courses | [Progress assessment] |
| Avg Attempts to Mastery | X.X | [Efficiency assessment] |
## Course-by-Course Analysis
### [Course Name] — [Status]
- **Progress**: X/Y resources completed (Z%)
- **Average Score**: X%
- **Strengths**: [Specific resources/topics mastered]
- **Growth Areas**: [Resources with low scores or high attempts]
- **Notable Patterns**: [From transcript analysis]
[Repeat for each enrolled course]
## Transcript Insights
Based on analysis of [N] lesson sessions:
- **Engagement Quality**: [High/Medium/Low] — [evidence: avg word count, questions asked, self-corrections]
- **Quiz Performance**: [First-attempt accuracy X%, common error patterns]
- **Scaffolding Dependency**: [Independent/Moderate/High] — [evidence: retry patterns, hint usage]
- **Misconceptions Identified**: [List any recurring wrong answers or misunderstandings]
## ZPD Classification
[Based on scores and patterns, classify where the student is operating:]
- Below 35%: Material is too challenging — recommend stepping back
- 36-69%: Productive struggle zone — maintain scaffolding
- 70%+: Ready to advance — fade scaffolds, increase challenge
**Current Zone**: [Classification with evidence]
## Recommendations
1. [Specific, actionable recommendation based on evidence]
2. [Another recommendation]
3. [Another recommendation]
## Early Warnings
[List any triggered warning indicators, or "No warnings at this time"]
- ⚠️ [Warning with evidence and recommended action]
# Course Performance Report: [Course Name]
**Date**: [Date]
**Enrolled Students**: [N]
**Course Type**: [Progression type]
## Class Overview
| Metric | Value |
|--------|-------|
| Average Score | X% |
| Completion Rate | X% |
| Students at Mastery (≥80%) | X/Y (Z%) |
| Students Needing Support (<70%) | X/Y (Z%) |
## Score Distribution
| Range | Count | Students | Recommended Action |
|-------|-------|----------|--------------------|
| 90-100% | X | [Names] | Recognition; extend with enrichment |
| 80-89% | X | [Names] | On track; maintain current approach |
| 70-79% | X | [Names] | Monitor; may need targeted support |
| 36-69% | X | [Names] | Active scaffolding; check understanding |
| 0-35% | X | [Names] | Urgent: reassess difficulty level |
## Per-Resource Analysis
| Resource | Avg Score | Completion | Avg Attempts | Problem Areas |
|----------|-----------|------------|-------------|---------------|
| [Name] | X% | X/Y | X.X | [Issues noted] |
[Flag resources where average score <70% — may indicate content issues rather than student issues]
## Students Needing Attention
### [Student Name] — [Score]%
- **Pattern**: [Brief description]
- **Recommended Action**: [Specific intervention]
## Course-Level Recommendations
1. [Recommendation about content/pacing/structure]
2. [Recommendation about specific resources]
3. [Recommendation about student grouping or differentiation]
# Intervention Alerts
**Date**: [Date]
**Scanned**: [N] active enrollments across [M] courses
## 🔴 URGENT — Immediate Action Needed
### [Student Name] — [Course Name]
- **Evidence**: [Specific data points]
- **Recommended Action**: [What to do]
## 🟡 WARNING — Monitor Closely
### [Student Name] — [Course Name]
- **Evidence**: [Specific data points]
- **Recommended Action**: [What to do]
## 🟢 POSITIVE — Recognition Opportunity
### [Student Name] — [Course Name]
- **Achievement**: [What they did well]
- **Recommended Action**: [How to reinforce]
When analyzing transcripts from get_transcript, look for these specific patterns:
lesson_feedback — these are submitted by the student after the lesson endsup (positive), down (negative), or null (skipped)slow_response, audio_quality, microphone_problems, network_issues, media_timing, quiz_problems, content_difficulty, bot_confusionquiz_response, freetext_response, or ordered_list_responseWhen writing reports:
npx claudepluginhub dselman/altistar-plugin --plugin altistarInterprets learning analytics data from LMS, quiz patterns, or engagement metrics and translates findings into actionable teaching decisions.
Audits course designs across 5 adversarial dimensions (alignment, evidence, cognitive load, persona simulation, prerequisite chains) and produces a confidence score. Runs in a clean-context sub-agent for unbiased synthesis.
Reviews learning progress across topics or specific ones, showing quiz scores, module completion, weak areas, trends, spaced repetition due items, and study recommendations.