skills/review/paper-reviewer/SKILL.md
# Scientific Paper Reviewer Skill (Multi-Venue) ## Quick Reference Card **START HERE**: Use the venue detection tree below before proceeding. ``` VENUE DETECTION TREE: User mentions "NeurIPS" OR has neurips template? → Use Section 3: NeurIPS Review Pathway User mentions "IEEE" OR has IEEE template/format? → Use Section 4: IEEE Review Pathway Cannot determine? → Ask user: "Is this for NeurIPS, IEEE, or another venue?" ``` --- ## 1. Skill Overview **Purpose**: Systematic, standards-comp
npx skillsauth add astoreyai/claude-skills skills/review/paper-reviewerInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
START HERE: Use the venue detection tree below before proceeding.
VENUE DETECTION TREE:
User mentions "NeurIPS" OR has neurips template?
→ Use Section 3: NeurIPS Review Pathway
User mentions "IEEE" OR has IEEE template/format?
→ Use Section 4: IEEE Review Pathway
Cannot determine?
→ Ask user: "Is this for NeurIPS, IEEE, or another venue?"
Purpose: Systematic, standards-compliant peer review for NeurIPS conferences and IEEE journals/transactions.
Key Features:
paper-reviewer-v2/
├── SKILL.md (orchestrator - this file)
├── components/
│ ├── shared/
│ │ ├── venue-detector.md ← START HERE
│ │ ├── paper-classifier.md
│ │ └── ethics-screener.md
│ ├── neurips/ ← NeurIPS-specific evaluation
│ └── ieee/ ← IEEE-specific evaluation
├── templates/
│ ├── neurips/review-synthesis.md
│ └── ieee/review-synthesis.md
└── reference/
├── neurips/[NeurIPS guidelines]
└── ieee/[IEEE guidelines]
Stage 1: Paper Classification (15 min)
components/shared/paper-classifier.mdStage 2: Technical Assessment (45-60 min)
components/neurips/technical-assessment.mdStage 3: Contribution Assessment (30-40 min)
components/neurips/contribution-assessment.mdStage 4: Clarity Assessment (20-30 min)
components/neurips/clarity-assessment.mdStage 5: Ethics & Checklist (20-30 min)
components/shared/ethics-screener.mdcomponents/neurips/checklist-processor.mdStage 6: Scoring & Synthesis (30-45 min)
templates/neurips/review-synthesis.md| Score | Label | Target % | Use When | |-------|-------|----------|----------| | 6 | Strong Accept | ~0.5% | Groundbreaking, flawless | | 5 | Accept | ~20-25% | Clear contribution, sound | | 4 | Borderline Accept | ~10% | Strengths > weaknesses | | 3 | Borderline Reject | ~10% | Weaknesses > strengths | | 2 | Reject | ~50-60% | Insufficient contribution | | 1 | Strong Reject | ~5% | Fundamental flaws |
SUMMARY: [100-200 words - factual overview]
STRENGTHS: [3-7 specific points with line refs]
1. [Strength with evidence]
2. [Strength with evidence]
...
WEAKNESSES: [3-8 specific points with line refs]
1. [Weakness with evidence and severity]
2. [Weakness with evidence and severity]
...
QUESTIONS: [3-5 questions for authors]
1. [Question] - Impact: [How answer affects evaluation]
2. [Question] - Impact: [How answer affects evaluation]
...
LIMITATIONS: [Yes/No + justification]
ETHICAL CONCERNS: [Yes/No + details if flagged]
SCORES:
- Quality: [1-4]
- Clarity: [1-4]
- Originality: [1-4]
- Significance: [1-4]
- Overall: [1-6] [Detailed justification]
- Confidence: [1-5] [Basis for confidence level]
Stage 1: Paper Classification (15 min)
components/shared/paper-classifier.mdStage 2: Technical Assessment (45-60 min)
components/ieee/technical-assessment.mdStage 3: Significance Assessment (30-40 min)
components/ieee/significance-assessment.mdStage 4: Clarity Assessment (20-30 min)
components/ieee/clarity-assessment.mdStage 5: Ethics Screening (15-20 min)
components/shared/ethics-screener.mdStage 6: Decision Recommendation (30-45 min)
templates/ieee/review-synthesis.mdDecision Tree:
Research technically sound?
├─ NO → REJECT (explain flaws)
└─ YES → Contribution significant?
├─ NO → REJECT (insufficient contribution)
└─ YES → Issues requiring changes?
├─ NONE → ACCEPT (rare on first submission)
├─ MINOR → MINOR REVISION
└─ MAJOR → MAJOR REVISION
SUMMARY: [Brief restatement of purpose/findings]
MAJOR CONCERNS:
1. [Methodological/analytical issues]
2. [Significant technical problems]
...
MINOR CONCERNS:
1. [Clarity/presentation issues]
2. [Editorial suggestions]
...
STRENGTHS:
- [Notable positive aspects]
- [Well-executed elements]
...
QUESTIONS FOR AUTHORS:
- [Clarifications needed]
- [Additional details required]
...
SUGGESTIONS FOR IMPROVEMENT:
- [Constructive, actionable feedback]
- [Reference additions with justification]
...
OVERALL ASSESSMENT: [Holistic evaluation]
RECOMMENDATION: [Accept / Major Revision / Minor Revision / Reject]
[Detailed justification for recommendation]
✓ Complete venue detection first
✓ Be thorough and systematic
✓ Provide specific, actionable feedback
✓ Maintain professional tone
✓ Support criticisms with clear reasoning
✓ Complete reviews promptly
✓ Protect confidentiality
✗ Skip venue detection ✗ Use AI tools to draft reviews ✗ Be vague or dismissive ✗ Speculate about authors ✗ Suggest excessive irrelevant references ✗ Review with conflicts of interest ✗ Share confidential information
| Aspect | NeurIPS | IEEE | |--------|---------|------| | Review Model | Double-blind | Single-blind (typical) | | Output | 1-6 score | Accept/Revise/Reject | | Page Limit | 9 pages strict | Flexible by journal | | Checklist | Mandatory | Not required | | Timeline | Conference cycle | ~90 days | | Acceptance Rate | ~25% | Varies by journal | | Emphasis | Novelty + reproducibility | Technical quality + impact | | Borderline Papers | Explicit 3-4 scores | Revision categories |
New users:
Experienced reviewers:
venue-detector.md - ALWAYS run firstpaper-classifier.md - Classification for both venuesethics-screener.md - Universal ethics checktechnical-assessment.md - Conference-focused evaluationcontribution-assessment.md - Novelty emphasisclarity-assessment.md - Reproducibility focuschecklist-processor.md - NeurIPS mandatory checklistscoring-calibration.md - 1-6 score determinationtechnical-assessment.md - Journal-depth evaluationsignificance-assessment.md - Practical impact focusclarity-assessment.md - Communication qualitydecision-recommendation.md - 3-tier decision logicCan't determine venue?
→ Use components/shared/venue-detector.md protocol
Unsure about scoring? → NeurIPS: Check acceptance rate targets in Section 3 → IEEE: Use decision tree in Section 4
Ethics concerns?
→ Flag immediately, consult ethics-screener.md
Paper spans multiple types?
→ Use paper-classifier.md to identify primary + secondary
Deadline pressure? → Prioritize technical assessment and decision, but complete all stages
Version: 2.0 (Multi-Venue) Supported: NeurIPS 2025, IEEE Journals/Transactions 2025 Last Updated: October 2025 Maintainer: Review Standards Committee
Changelog:
Confidentiality: Never share or use unpublished work Objectivity: Review based on merit, not personal bias Constructiveness: Help authors improve their work Professionalism: Maintain respectful, courteous tone Integrity: Report suspected misconduct promptly Timeliness: Complete reviews within expected timeframe Independence: Declare conflicts, decline if present
END OF MAIN SKILL DOCUMENTATION
For detailed component usage, see individual files in components/, templates/, and reference/ directories.
tools
# YouTube Transcriber Pipeline - Claude Code Skill **Version**: 1.0.0 | **Status**: Ready for Claude Code Integration ## Overview Complete 4-skill pipeline for extracting, formatting, organizing, and archiving YouTube transcripts. Integrates with Claude Code CLI using the system's configured API keys (no manual key management needed). ## Capabilities This skill provides a complete workflow: 1. **Extract Facts** - AI-powered fact extraction from transcripts (uses Claude API via Claude Code)
content-media
# YouTube Transcriber Skill Extract transcripts from YouTube videos, playlists, and channels with automatic intelligent processing. ## Overview One unified command that intelligently assesses input and handles everything—single videos, batch files, playlist expansion, channel extraction. No need to choose between commands; it figures out what to do. ## Capabilities - **Auto-Detect Input**: Single URL, file of URLs, playlist, channel - **Smart Expansion**: Automatically expands playlists/cha
development
# Trading Analysis Skill **Version**: 1.0.0 **Category**: Financial Analysis / Trading **Author**: Claude Code **Last Updated**: November 22, 2025 ## Overview Comprehensive trading performance analysis and edge identification system for Interactive Brokers accounts. Analyzes CSV statements to identify trading patterns, position sizing issues, time-of-day edges, and risk management problems. ## Features ### 1. **CSV Statement Parsing** - Parse Interactive Brokers activity statements (CSV for
development
# System Health Check & Cleanup Skill **Version**: 1.0.0 **Category**: System Administration / Performance Optimization **Author**: Claude Code **Last Updated**: November 22, 2025 ## Overview Automated system health monitoring and cleanup workflow. Diagnoses performance issues, identifies resource bottlenecks, fixes orphaned services, kills stale processes, and cleans cache bloat. Designed for archimedes (32c/125GB) but works on any Linux system. ## Features ### 1. **System Diagnostics** -