exam-forecast

Analyzes past law school exams to forecast subject weighting and question patterns for upcoming exams.

Updated Jun 17, 2026
One-click install
npx skills add https://github.com/tk1cntt/PhapChe --skill exam-forecast-tk1cntt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: exam-forecast
Source: https://github.com/tk1cntt/PhapChe/tree/main/docs/claude-for-legal-main/law-student/skills/exam-forecast
Command: npx skills add https://github.com/tk1cntt/PhapChe --skill exam-forecast-tk1cntt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Law students often study every topic equally because they cannot tell which subjects their professor actually emphasizes on exams. This Skill analyzes past exams from the same professor to surface recurring patterns — subject weighting, question styles, recurring traps, and policy-vs-doctrine mix — so study time can be weighted toward what is historically tested. ## Core Features & Use Cases - Past Exam Pattern Analysis: Reads each past exam for format, subject coverage, question style, fact-pattern density, and recurring traps, then rolls up stable versus variable patterns across exams. - Weighted Forecast Output: Produces a dated forecast file with a subject-weighting table, question-style forecast, professor hobby horses, and a heavy/moderate/sanity-check study emphasis recommendation. - Confidence Discipline: Flags thin samples (fewer than 3 exams), marks inferences as uncertain, and falls back to syllabus-only coverage when no past exams exist. - Use Case: A student with four past Contracts exams from the same professor shares them before finals and receives a forecast showing that consideration and modification consistently account for 30% of points, so they allocate review time accordingly. ## Quick Start Analyze these past exams from my Contracts professor and forecast what topics and question formats are likely on my upcoming final.

Frequently Asked Questions about exam-forecast

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict what will be on my law school exam?

Share past exams from the same professor and the skill analyzes subject weighting, question styles, and recurring traps across them. It produces a forecast framed as a study-time weighting heuristic, not a prediction of specific questions.

How many past exams do I need for an exam forecast?

Three or more past exams give moderate confidence; six or more is strong. With only one or two exams the sample is flagged as thin, and any inferred pattern is treated as noise rather than signal.

Can exam pattern analysis work without past exams from my professor?

No. Without past exams from the same professor the skill cannot forecast and falls back to listing what the current syllabus covers. Same-professor exams are the highest-signal input for pattern analysis.

What input formats does the exam analysis accept?

Past exams can be provided as PDFs, pasted text, or file paths. If the exams include the professor's name it is used to match patterns; otherwise matching falls back to subject and structure.

What are the limitations of past-exam forecasting?

The forecast cannot predict specific questions, account for professor changes you have not shared, or justify skipping syllabus topics. It weights study time based on historical patterns and explicitly flags uncertain inferences.