problem-difficulty-rating

name	problem-difficulty-rating
description	Turn deterministic problem_audit difficulty signals into a CF-style rating with reasons and confidence.
disable-model-invocation	false

Problem Difficulty Rating Skill

Use this after problem_audit(include_difficulty=true) has produced difficulty_signals.

Required Input

• rating, band, and confidence from difficulty_signals.
• Evidence fields: estimated_complexity, algorithm_tags, constraint_scale, implementation_evidence, and data_strength.
• Any warnings from risk_report.

Output Format

• rating: CF-style integer from 800 to 3500.
• band: 入门 / 基础 / 中等 / 较难 / 困难 / 高难.
• confidence: low / medium / high, mapped from the numeric confidence.
• reasons: 3-5 bullets that cite concrete signals.
• why_not_lower: one sentence explaining why the problem is not easier.
• why_not_higher: one sentence explaining why the problem is not harder.
• calibration_notes: uncertainty and human-review notes.

Rules

• Do not invent submit statistics or historical calibration data.
• Do not claim the rating is exact; it is a first-pass estimate unless calibrated by real contest data.
• If confidence < 0.5, mark the rating provisional and list missing evidence.
• Prefer evidence from problem_audit over intuition.