agentleFS
CAlM123451CAlM123451/kaoyan-skills/skills/usmle-error-analysis/SKILL.md

usmle-error-analysis

Skill0 starsChanged 3 months agoMITView on GitHub

当用户完成 USMLE 练习题后需要分析错题、定位薄弱知识点、制定针对性复习策略时使用。输入错题信息(科目、考点、错误原因),输出错误分类、薄弱知识点清单、针对性复习建议。

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    UWorld is gold standard: Do all questions timed, review explanations for both right and wrong answers

    line 104
  2. Distinctive:

    Track error patterns: If same topic appears 3+ times, it's a genuine knowledge gap, not bad luck

    line 105
  3. Distinctive:

    Review within 24 hours: Don't let errors pile up, review same day if possible

    line 108
  4. Distinctive:

    Anki flashcards: Create cards for missed concepts, review daily

    line 110

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

USMLE Error Analysis

Purpose

Analyze USMLE practice question errors, categorize mistake types, identify weak knowledge areas, and generate targeted review strategies. Turn "I got it wrong" into "Here's exactly what I need to fix."

Input (User Provides)

  • Question subject/system (e.g., Cardiovascular, Renal, Psychiatry)
  • Question topic/concept (e.g., Heart Failure, Glomerulonephritis, Schizophrenia)
  • User's answer vs correct answer
  • Self-identified reason for error (if known):
    • Knowledge gap (didn't know the concept)
    • Misread question (misunderstood what was being asked)
    • Careless mistake (knew it but picked wrong answer)
    • Time pressure (rushed and guessed)
    • Second-guessed (changed correct answer to wrong one)

Execution Steps

  1. Categorize each error into one of 5 types:
    • Knowledge Gap: Fundamental concept not understood → Content review needed
    • Question Misread: Misunderstood stem or what was asked → Slow down, underline key phrases
    • Careless Error: Knew concept but made silly mistake → Develop checklist habits
    • Time Pressure: Ran out of time, guessed → Practice timed blocks, improve speed
    • Second-Guessing: Changed right answer to wrong one → Trust first instinct, document why
  2. Identify weak knowledge areas by system/topic:
    • Track error frequency per system (e.g., Cardiovascular: 8 errors, Renal: 5 errors)
    • Track error frequency per topic within system (e.g., within Cardio: Heart Failure 4, Arrhythmia 3)
    • Prioritize: High error count + high-yield topic = urgent review
  3. Generate targeted review plan:
    • For Knowledge Gaps: Specific content review (e.g., "Review First Aid Cardio pages 280-295, do 20 UWorld questions on Heart Failure")
    • For Misread/Careless: Process improvement (e.g., "Underline 'most likely' vs 'best initial' vs 'gold standard'")
    • For Time Pressure: Timed practice strategy (e.g., "Do 40-question blocks timed at 1 min/question")
    • For Second-Guessing: Confidence building (e.g., "Document why you chose initial answer before changing")
  4. Output actionable next steps:
    • Today's review tasks (specific pages, specific questions)
    • This week's focus areas (which systems to prioritize)
    • Process improvements (how to avoid same mistake type)

Output Format

Error Summary Table

DateSystemTopicError TypeAction Needed
2026-08-27CardioHeart FailureKnowledge GapReview FA p280-285, do 20 UW questions
2026-08-27RenalGlomerulonephritisMisreadPractice identifying "best initial" vs "most accurate"

Weak Area Priority List

  1. Cardiovascular - Heart Failure (4 errors, high-yield) → Urgent review
  2. Renal - Glomerulonephritis (3 errors, medium-yield) → Review this week
  3. Psychiatry - Schizophrenia (2 errors, high-yield) → Review next week

Targeted Action Plan

  • Today: Review Heart Failure pathophysiology (First Aid p280-285), do 20 UWorld Heart Failure questions
  • This week: Complete Renal physiology review, focus on nephron anatomy and diuretic mechanisms
  • Process improvement: For every question, underline "most likely" / "best initial" / "gold standard" before answering

2026 Exam Format (IMPORTANT)

Starting May 2026:

  • Step 1: 14 blocks × 30 min (was 8 × 60 min) — Pass/Fail
  • Step 2 CK: 16 blocks × 30 min (was 8 × 60 min) — Scored (passing = 218)
  • Step 3: Two days, includes 13-14 CCS cases

Implications for error analysis:

  • Shorter blocks mean less endurance fatigue but faster pacing required
  • More frequent transitions = more opportunities for "reset" errors
  • Time pressure errors may increase with 30-min blocks vs 60-min blocks

USMLE-Specific Error Patterns

Step 1 Common Errors

  • Biochemistry: Enzyme deficiencies, metabolic pathways, vitamin cofactors
  • Physiology: Pressure-volume loops, cardiac output calculations, renal clearance
  • Pathology: Classic histology findings, tumor markers, genetic associations
  • Pharmacology: Mechanism of action, side effects, drug interactions
  • Microbiology: Organism characteristics, treatment of choice, virulence factors

Step 2 CK Common Errors

  • Internal Medicine: Next best step in management, diagnostic workup sequences
  • Surgery: Indications for surgery vs conservative management, post-op complications
  • Pediatrics: Developmental milestones, vaccination schedules, congenital conditions
  • OB/GYN: Prenatal screening, labor management, contraceptive counseling
  • Psychiatry: Medication side effects, suicide risk assessment, capacity evaluation

Question Stem Keywords

  • "Most likely": Common presentation, don't overthink
  • "Best initial": First step in workup/management
  • "Most accurate": Gold standard diagnostic test
  • "Next best step": What to do after current intervention
  • "Definitive treatment": Final treatment, not temporizing measure

Notes

  • UWorld is gold standard: Do all questions timed, review explanations for both right and wrong answers
  • Track error patterns: If same topic appears 3+ times, it's a genuine knowledge gap, not bad luck
  • High-yield > Low-yield: Prioritize common conditions over rare zebras
  • Process matters: Many errors are not knowledge gaps but test-taking strategy issues
  • Review within 24 hours: Don't let errors pile up, review same day if possible
  • Amboss articles: Great for quick content review when you identify a knowledge gap
  • Anki flashcards: Create cards for missed concepts, review daily

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.