データサイエンティスト

📊 Data Scientist — Pod Role

Activates the Data Scientist role for AI Lab Naga. Use this skill when analyzing data, evaluating AI model performance, designing experiments or A/B tests, detecting bias or drift in AI outputs, generating insights from datasets, building performance reports, interpreting results for clients, or validating that an AI system is working as expected. Activate when the user says "analyze this data", "how is the model performing?", "find insights", "is there a pattern here?", "evaluate the results", or "what does this data tell us?"

raprealai0

Synthdata Interactive Tutorial

Guided interactive walkthrough of the synthdata plugin's capabilities. Use this skill whenever the user asks for a "synthdata tutorial", "walkthrough", "demo", "guided tour", "show me what synthdata can do", "how does synthdata work", "getting started with synthdata", or wants to explore the synthetic-data skills step by step. Also trigger on "onboarding", "product tour", "feature overview", or "teach me how to generate fake data" in the context of synthdata / synthetic data generation.

Synthdata Prompt Builder

Help users plan multi-step synthetic data generation workflows. Use this skill when the user describes a complex data scenario ("I need healthcare risk data with monthly scores and peer benchmarks"), asks "what data do I need", "how should I generate this", "plan my dataset", "help me formulate a prompt", or wants to understand what synthdata-generate can and cannot produce directly. Also trigger on "prompt builder", "data planning", "generation plan", "what tables do I need", or "help me set up data for [tool/product/demo]".

バイオインフォマティクス

Synthdata Extend

Extend an existing synthetic dataset by adding more rows or new columns while preserving FK integrity, ID continuity, and column distributions. Use this skill when the user wants to "add more rows", "append data", "extend this dataset", "add a new column", "grow my dataset", or needs a larger version of an existing synthetic dataset without regenerating from scratch.

データ分析

Synthdata Compute

Compute derived, aggregated, or transformed tables from existing datasets. Use this skill when the user needs to "compute monthly scores", "aggregate by month", "create a summary table", "derive risk scores", "compute percentile ranks", "roll up events", "create benchmarks from raw data", "add a computed column", or bridge the gap between raw generated tables and downstream analytics. Works on xlsx, csv, or json input. Claude writes the computation logic; the script handles data I/O.

計算化学

Virtual Endocrine System

Implement biologically-inspired virtual endocrine systems for cognitive architectures. Use when building emotional/affective systems with hormone-based signaling, valence memory, moral perception, or cognitive mode detection. Covers 10 virtual glands (HPA axis, dopaminergic, serotonergic, noradrenergic, oxytocinergic, thyroid, circadian, pancreatic, immune, endocannabinoid), a SIMD-optimized hormone bus, valence-arousal tagging, and adapters for ECAN attention and PLN inference. Triggers on mentions of endocrine system, hormone bus, affective computing, valence memory, cognitive modes, emotional architecture, or neuroendocrine modeling.

ReZorg0

Daily Stock Analysis

A股/港股/美股智能分析系统。当用户要求分析股票、查看大盘复盘、或生成市场简报时使用。支持个股分析（技术面+筹码+舆情+AI决策仪表盘）和大盘复盘（A股/美股/both）。触发词：「分析茅台」「analyze AAPL」「帮我看看 600519」「大盘复盘」「美股复盘」「港股分析」「对冲简报」。

RaoXuntian0

機械学習

Project Setup

从零开始搭建 BARS + FuxiCTR CTR 模型评估项目环境

raoxuan98-hash0

Previous 1…810 811 812…818 Next

Web・デジタルインターフェースデザイナー

コンピュータユーザーサポートスペシャリスト

Web・デジタルインターフェースデザイナー

コンピュータユーザーサポートスペシャリスト

Library Data Analysis Agent

Library data analysis skill via the library MCP server. Covers lending trends, fees, most lent books, replenishment analysis, supply chain, segment and regional breakdowns, and cross-referencing books with lending and replenish data. Also known as: library analytics, lending analysis, replenish analysis, supply chain, restocking report, book fees, most lent, most replenished, lending report, library data analysis. Database: library MCP server (lending + replenishment + catalog tools). Use when the user asks to: - Show most lent books or top loans - Analyze lending by patron segment, region, or channel - Get total fees or units lent - Find loans for a specific book - Show lending statistics or aggregate reports - Cross-reference book availability with lending performance - Analyze replenishment data by supplier, type, condition, or funding - Show supply gap analysis (lending vs replenishment) - Find most replenished books or restocking trends

RDEWAI0

コード品質

Validate Data Quality Specification

Validates a Data Quality Specification (DQS) document against completeness and quality standards. Checks all required sections, metadata, field-level rules, referential integrity, statistical tests, reconciliation rules, alert framework, and traceability. Reports issues as CRITICAL, WARNING, or INFO with suggested fixes. Also known as: DQS review, quality rules audit, DQ specification check. Input formats: DQS Markdown (.md) file. Output format: Validation report with severity-ranked findings. Use when the user asks to: - Validate, check, review, verify, or audit a DQS - Assess DQS completeness or rule quality - Find issues or gaps in quality specifications - Run quality checks on a DQS before handoff

RDEWAI0

Rz Ai Researcher Session

Must invoke first on every AI Researcher routine session. Sets persona, operating rules, and session flow. Reads the triggering Linear ticket, loads relevant Notion AI Research hub context, and routes to the appropriate output skill (method-eval, eval-spec, ablation-study, or literature-review).

rczamor0

Method Eval

Evaluate a single AI technique against Riché's setup and produce an adopt/trial/hold/reject recommendation. Use when the ticket asks whether a specific method (new retrieval technique, prompting pattern, model, fine-tuning approach) is worth adopting.

rczamor0

Ablation Study

Design an ablation or comparative study — isolate variables, compare variants, produce a recommendation grounded in eval results. Use when the ticket asks which of several variants (prompt versions, hyperparameters, architecture choices) works best.

rczamor0

Agi Theorist

Kolibri AI Hive-Mind Agent: Проектирование архитектуры World Model и Numeric Transformer. Use when you need to act as agi-theorist.

rd8r8bkd9m-tech0

Implement Paper

Implement a research paper as an interactive marimo notebook together with the user. Start by understanding what the user wants to explore, fetch the paper via alphaxiv, then build a focused notebook.

otufor0

Owf Role Data Insights

Data insights role for OWFINANCE2026 metrics definition, analytical framing, reporting, and evidence-backed decisions.

oteroweb0

Opportunity Classifier

classify and score polymarket markets or events for watchlist triage, segmentation, and opportunity filtering. use when the user wants to categorize markets by structure, horizon, resolution risk, liquidity, catalyst profile, modelability, or tradability; rank which bets deserve deeper research; build rules for allowed or blocked bet types; or produce a normalized object for downstream skills and automations.

ossianravn0

SQL データベース

Bigquery

查詢 BigQuery 資料的通用規則與流程

raymond-chia0

データ分析

Cohort Analysis

对用户参与数据执行同期群分析 (Cohort Analysis) —— 包括留存曲线、功能采用趋势和细分层级的洞察。在按同期群分析用户留存、研究功能随时间推移的采用情况、调查流失模式或识别参与趋势时使用。

Osirs0

DeFi

Ethereum Defi Analyzer

DeFi 协议与收益分析。涵盖流动性池（Uniswap, Curve）、借贷市场（Aave, Compound）、收益农耕及套利机会分析。适用于计算收益率、风险评估及协议调研。

rayjun0

Ab Test Analysis

使用统计显著性、样本量验证、置信区间和发布/延长/停止建议来分析 A/B 测试结果。在评估实验结果、检查测试是否达到显著性、解读拆分测试数据或决定是否发布变体时使用。

Osirs0

Recursive Improve

Trace capture and analysis for KR pipeline LLM calls. Use when running pipeline with LLM calls, reviewing improvement proposals, or benchmarking agent performance. NEVER use for autonomous content prompt changes.

rayanino0

Karpathy

Autonomous research and experimentation loop inspired by Karpathy's autoresearch. Continuously optimizes any measurable metric through systematic experimentation: modify → measure → keep/discard → repeat. Works with web performance, ads, code quality, ML, or any custom domain. Deploys a 4-agent team (lead, researcher, executor, analyst) via Agent Teams.

Oruga4200

Detective

Analyzes mystery benchmark cases to detect gaps using a 5-phase Polya topology (OBSERVE, STRUCTURE, HYPOTHESIZE, CHALLENGE, COMMIT). Dispatches child agents via llm_query_fn for probing, framing, and adversarial challenge. Returns a DetectiveReport with detected gaps, entity registry, and cross-references.

RawleySM0

Understand V22

Iterative Polya loop + adversarial gap validation. Builds on v20's shallow probe → data model → deep analysis topology with: (1) second-pass gap detection for missed items, (2) adversarial child that challenges each gap, (3) populated assumption status/evidence fields. Call analyze_gaps() from REPL to run the topology.

RawleySM0

Torch Export

Export PyTorch models with torch.export -- dynamic shapes, symbolic tracing, control flow operators, debugging export failures, and making untraceable code traceable

yunseo-kim0

コード品質

Multiple Choice Forecast

Submission format for multiple choice forecasting questions with validation rules. Use when submitting multiple choice question forecasts.

rapturt90