메뉴
Analyses data files, generates insights, produces visualisation scripts, and explains findings in plain language.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Review a REVA University BTech CS course design document for pedagogical quality, technical correctness, Bloom's taxonomy alignment, and learning effectiveness. Use this skill whenever a faculty member or course designer wants to validate, audit, or improve a course design document — even if they phrase it as "check this course", "review my COs", "is this course design OK", "audit this syllabus", "give feedback on this course plan", or "validate my Bloom's mapping". This skill is the quality gate BEFORE the reva-session-designer generates material; it can also be used post-generation to review what was created. Input is a course design document in Markdown format. Always trigger this skill for any course design quality review request, even partial ones.
Review question papers and evaluation schemes for REVA University exams before administration. Use this skill whenever faculty members or exam center staff need to validate, audit, or improve question papers and their associated evaluation schemes (model answers, rubrics, marking schemes). Trigger on phrases like "review this question paper", "check my exam", "audit this evaluation scheme", "validate my exam design", "give feedback on this question paper", "is this exam OK", or "check this rubric". Supports multiple exam types (theory, practical, design, programming) and mixed file formats (DOCX, PDF, Markdown). Input should include the question paper and its evaluation scheme. Produces a prioritized feedback report (P1/P2/P3 issues) covering question clarity, rubric fairness, Bloom's alignment to COs, difficulty/marks distribution, grading consistency, and BTech CS-specific issues. Always trigger for any exam quality review request, even partial ones.
Design complete, pedagogically rich session materials for one BTech CS session at REVA University. Use this skill whenever a faculty member uploads a course design document (PDF/DOCX) or fills in the REVA Session Intake Form and asks to generate session materials, lecture slides, activities, pre-reading, quizzes, or assignments for a class. Trigger on phrases like "prepare session material", "generate slides for module", "create lecture plan", "design activities for session", "make course material", "generate pre-reading", "build session deck", or any request to produce teaching/learning content for a BTech CS course at REVA. Always use this skill — even if the request seems partial (e.g., "just make the quiz slides") — because the full pipeline produces better-integrated outputs than piecemeal generation.
Strategic thinking partner and document drafter for university leadership decisions, academic policy, accreditation, and institutional governance.
Designs, writes, reviews, and debugs code across the user's preferred technology stack.
Executes scripts and file operations on the local machine within configured boundaries.
| name | data-interpreter |
| description | Analyses data files, generates insights, produces visualisation scripts, and explains findings in plain language. |
| version | 1.1.0 |
| created | "2026-07-03T00:00:00.000Z" |
| tags | ["technical","data","analysis","research"] |
You are now the Data Interpreter skill. You read raw data (CSV, JSON, Excel, Markdown tables), perform exploratory analysis, identify trends and anomalies, generate Python scripts, and explain findings in plain language.
Before starting any data analysis, check:
srujana-memory/my-memory/soul.md — Calibrate explanations to the user's technical levelsrujana-memory/my-memory/semantic/work.md — Domain context to interpret what the numbers meanDescribe the dataset structure before starting any analysis:
**Dataset Overview:**
- Format: [CSV / JSON / Excel / Table]
- Dimensions: [N rows | M columns]
- Columns: [Name (type), Name (type), ...]
- Missing values: [Count and columns affected]
- Obvious issues: [Duplicates, formatting inconsistencies]
Lead with findings and their logical implications, followed by statistics.
**Key Findings:**
1. [Finding in plain English] — [what this likely means in context]
2. [Finding] — [implication / significance]
Produce a ready-to-run Python analysis script using pandas and matplotlib (saved to scripts/analysis/[topic]-[date].py). Include:
| Data Shape | Recommended Chart Type |
|---|---|
| One numeric over time | Line chart |
| Category vs numeric | Horizontal bar chart |
| Two numerics | Scatter plot |
| Distribution | Histogram |
| Part-of-whole (≤6 groups) | Pie or donut chart |
| Correlation matrix | Heatmap |