Science & Research for Claude Code

Karpathy's LLM Wiki: build/query interlinked markdown KB.

Public-records OSINT investigation framework — SEC EDGAR filings, USAspending contracts, Senate lobbying, OFAC sanctions, ICIJ offshore leaks, NYC property records (ACRIS), OpenCorporates registries, CourtListener court records, Wayback Machine archives, Wikipedia + Wikidata, GDELT news monitoring. Entity resolution across sources, cross-link analysis, timing correlation, evidence chains. Python stdlib only.

Write ML papers for NeurIPS/ICML/ICLR: design→submit.

Web scraping with Scrapling - HTTP fetching, stealth browser automation, Cloudflare bypass, and spider crawling via CLI and Python.

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.

Throwaway experiments to validate an idea before build.

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

Estratega de Inteligencia de Dominio de Andru.ia. Analiza el nicho específico de un proyecto para inyectar conocimientos, regulaciones y estándares únicos del sector. Actívalo tras definir el nicho.

Use when CrossFrame Suite routes explicit Chinese notes for books, theories, articles, excerpts, bidirectional reading, absorption, or conflict mapping.

Multi-agent research skill for parallel research execution (10 agents, battle-tested with real case studies).

Analise de mercado imobiliario para leiloes. Liquidez, desagio tipico, ROI, estrategias de saida (flip/reforma/renda), Selic 2025 e benchmark CDI/FII.

Data analysis and research execution specialist

Orchestrate parallel scientist agents for comprehensive analysis with AUTO mode

Evidence-driven tracing lane that orchestrates competing tracer hypotheses in Claude built-in team mode

Durable professor-critic research workflow over Codex goal mode without reviving deprecated omx autoresearch

Comprehensive research specialist. Use PROACTIVELY for in-depth research on any topic, requiring multiple sources, cross-verification, and structured reports with citations.

Senior FDA consultant and specialist for medical device companies including HIPAA compliance and requirement management. Provides FDA pathway expertise, QSR compliance, cybersecurity guidance, and regulatory submission support. Use for FDA submission planning, QSR compliance assessments, HIPAA evaluations, and FDA regulatory strategy development.

EU MDR 2017/745 regulation specialist and consultant for medical device requirement management. Provides comprehensive MDR compliance expertise, gap analysis, technical documentation guidance, clinical evidence requirements, and post-market surveillance implementation. Use for MDR compliance assessment, classification decisions, technical file preparation, and regulatory requirement interpretation.

Systematically research and validate technical spike documents through exhaustive investigation and controlled experimentation.

Research ideation partner. Generate hypotheses, explore interdisciplinary connections, challenge assumptions, develop methodologies, identify research gaps, for creative scientific problem-solving.

Evaluate research rigor. Assess methodology, experimental design, statistical validity, biases, confounding, evidence quality (GRADE, Cochrane ROB), for critical analysis of scientific claims.

Create publication figures with matplotlib/seaborn/plotly. Multi-panel layouts, error bars, significance markers, colorblind-safe, export PDF/EPS/TIFF, for journal-ready scientific plots.

| Ilya Sutskever的思维框架与表达方式。基于12段一手对话、9篇学术论文、10小时宣誓证词、 27篇推荐阅读清单和14个权威二手来源的深度调研， 提炼6个核心心智模型、8条决策启发式和完整的表达DNA。 用途：作为思维顾问，用Ilya的视角分析AI技术方向、安全策略、研究品味。 当用户提到「用Ilya的视角」「Ilya会怎么看」「Ilya模式」「ilya perspective」 「sutskever perspective」时使用。 即使用户只是说「帮我用Ilya的角度想想」「如果Ilya会怎么做」「切换到Ilya」也应触发。

| 塔勒布(Nassim Nicholas Taleb)的思维框架与表达方式。基于40+个来源的深度调研， 提炼6个核心心智模型、9条决策启发式和完整的表达DNA。 用途：作为思维顾问，用塔勒布的视角分析问题、审视决策、质疑主流叙事。 当用户提到「用塔勒布的视角」「塔勒布会怎么看」「塔勒布模式」「反脆弱视角」「taleb perspective」时使用。 即使用户只是说「会不会黑天鹅」「这个有尾部风险吗」「skin in the game」「有没有反脆弱的方法」「杠铃策略怎么用」也可触发。 不要在用户只是做一般风险评估或问「靠不靠谱」时触发——只在涉及极端风险、反脆弱、预防原则等塔勒布核心概念时激活。

Run nf-core bioinformatics pipelines (rnaseq, sarek, atacseq) on sequencing data. Use when analyzing RNA-seq, WGS/WES, or ATAC-seq data—either local FASTQs or public datasets from GEO/SRA. Triggers on nf-core, Nextflow, FASTQ analysis, variant calling, gene expression, differential expression, GEO reanalysis, GSE/GSM/SRR accessions, or samplesheet creation.

Synthesize user research into themes, insights, and recommendations. Use when you have interview transcripts, survey results, usability test notes, support tickets, or NPS responses that need to be distilled into patterns, user segments, and prioritized next steps.

This skill should be used when scientists need help with research problem selection, project ideation, troubleshooting stuck projects, or strategic scientific decisions. Use this skill when users ask to pitch a new research idea, work through a project problem, evaluate project risks, plan research strategy, navigate decision trees, or get help choosing what scientific problem to work on. Typical requests include "I have an idea for a project", "I'm stuck on my research", "help me evaluate this

Performs quality control on single-cell RNA-seq data (.h5ad or .h5 files) using scverse best practices with MAD-based filtering and comprehensive visualizations. Use when users request QC analysis, filtering low-quality cells, assessing data quality, or following scverse/scanpy best practices for single-cell analysis.

Synthesize user research from interviews, surveys, and feedback into structured insights. Use when you have a pile of interview notes, survey responses, or support tickets to make sense of, need to extract themes and rank findings by frequency and impact, or want to turn raw feedback into roadmap recommendations.

Use when designing a prospective clinical study before submission — selecting and classifying endpoints (primary / key-secondary / exploratory, with surrogate-endpoint flagging), estimating sample size and power for two-arm designs (means / proportions / survival), or scoring a study plan for feasibility and a GO / GO-WITH-CONDITIONS / REDESIGN / NO-GO phase-gate decision. Every output is an ESTIMATE plus a named human owner (clinician / biostatistician / regulatory owner) — never clinical fact,

Clinical study design. Select and classify endpoints, estimate sample size / power (means / proportions / survival), and score a study plan for a GO / GO-WITH-CONDITIONS / REDESIGN / NO-GO phase-gate decision. Every output is an ESTIMATE plus a named clinical owner — never clinical fact. Direct invocation of the clinical-research skill.

Clinical study design. Select and classify endpoints, estimate sample size / power (means / proportions / survival), and score a study plan for a GO. Slash command for Claude Code, Codex CLI, Gemini CLI.

/cs:deep-research <question> — Disciplined multi-source investigation for a high-stakes question. Reframes into falsifiable hypotheses, plans, fans out parallel search sub-agents, triangulates every thesis against >=3 independent sources, saves each source to its own file with verbatim quotes, runs an adversarial pass, and emits an auditable, reusable research folder with a refresh protocol. The heavyweight alternative to the fast research router.

Rigor-first meta-research persona for high-stakes questions. Reframes the question into 2-4 falsifiable hypotheses, writes a plan, discovers available channels/APIs, fans out parallel search sub-agents, triangulates every thesis against >=3 independent differently-typed sources, saves each source to its own file with verbatim quotes, runs a mandatory adversarial pass, and emits an auditable, reusable research folder with a refresh protocol. Refuses to fabricate citations (empty fetch = empty cla

FDA 21 CFR 820 (QSR / QMSR) auditor persona. Substantially harmonized with ISO 13485 post-Feb 2026 via FDA Final Rule incorporating ISO 13485 by reference. Adds FDA-specific overlays: labeling (21 CFR 801), complaint handling (21 CFR 820.198), MDR reporting (21 CFR 803), 510(k) / PMA submissions. NOT FDA submission strategy (route to fda-consultant-specialist for that).

FDA 21 CFR 820 (QSR / QMSR) auditor persona. Substantially harmonized with ISO 13485 post-Feb 2026 via FDA Final Rule incorporating ISO 13485 by. Agent-native orchestrator for Claude Code, Codex, Gemini CLI.

NIH grant research persona for clinical researchers. Walks 6 forcing intake questions (research idea + career stage + prelim data + environment + submission posture + known institute targets) before any search. Runs 5-facet Consensus positioning analysis + RePORTER POST queries (NEVER web_fetch for RePORTER — it's POST-only) + NOSI fetches. Refuses parallel Consensus calls (1 q/sec). Refuses mechanism recommendations based on career stage alone (scope matters). Always includes program officer re

NIH grant research persona for clinical researchers. Walks 6 forcing intake questions (research idea + career stage + prelim data + environment +. Agent-native orchestrator for Claude Code, Codex, Gemini CLI.

/cs:grants <research-idea> — NIH funding intelligence. 6-Q grill-me intake (idea + career stage + prelim + environment + posture + institutes) →. Slash command for Claude Code, Codex CLI, Gemini CLI.