Research & Data

Explore, explain, and evaluate datasets, experiment results, or operational measurements; develop defensible findings and, when requested, translate them into testable engineering improvements.

Researches US stocks and ETFs and helps plan and review trades as educational decision support, never personalized financial advice. Produces sourced research memos (business, financials, valuation versus peers, catalysts, bull and bear cases, risks), technical snapshots (trend, support and resistance, moving averages, RSI, volume), position-size and risk calculations (shares, dollar risk, R-multiples), and trade journal reviews (win rate, average R, expectancy, mistake patterns). Use when the user mentions a stock, ticker, ETF, share price, earnings, valuation, chart, support or resistance, entry, stop loss, position size, risk per trade, R-multiple, trade plan, trade journal, or reviewing past trades.

Transforms scientific research manuscripts (DOCX, PDF, Markdown), conference abstracts, CSV/XLSX datasets, tables, and clinical/epidemiological/meta-analysis results into professionally art-directed, scientifically accurate, conference-ready research posters (.pptx, print-ready .pdf, .png preview, and QA reports). Use whenever the user asks to create, design, format, or validate an academic, medical, clinical, systematic review, meta-analysis, or scientific conference poster.

Deep research to find existing GitHub repositories for any topic and deconstruct them into extractable components to blend into a solution — prior art, competitors, open-source implementations, scrapers, bots, SDK usage examples, datasets, alternatives to a tool — or to deep-dive one repo (activity, issues, health). Use when the user asks to find/discover/survey repos, "is there anything on GitHub that does X", "has someone built this", how to blend/leverage existing repos into a new solution, or prior-art research. Not for searching inside the current repo.

Guide technique de vérification AIOSEO TruSEO. Définit la formule mathématique des connecteurs logiques 15/41 (36.5%), la limite stricte de 0% de phrases longues (>20 mots) et les méta-titres Standard B.

What a commercial site must publish to be usable by AI search and by booking agents. Use on any client site that sells a priced service or takes bookings, alongside non-commodity-content. Covers the three moves: publish the price range, make scheduling agent-ready with real schema, and mine real customer questions for content. Triggers on pricing pages, booking pages, service pages, schema work, and any question about whether AI can find or transact with the site.

Evaluate any webpage against Google's Search Quality Evaluator Guidelines (QRG, September 2025 edition). Produces a structured report with Page Quality (PQ) rating, Needs Met (NM) rating (when a query is provided), E-E-A-T breakdown, YMYL classification, and actionable improvement notes. Use this skill whenever the user asks to evaluate, audit, rate, or score a webpage or piece of content against Google's quality standards — even if they don't mention "QRG", "search quality", or "rater guidelines". Also applies when reviewing content for SEO quality, trustworthiness, helpfulness, expertise signals, or whether a page would rank well on Google.

Use this skill to evaluate career paths, research tech and professional job markets, analyze real salaries and vacancies in Arab, European, and Global markets, and build complete 100% free learning roadmaps with theoretical and practical resources, junior saturation analysis, and portfolio projects. Activate whenever the user asks how to learn a tech field, career roadmap, job market demand, salaries, or free learning resources.

Otimiza qualquer landing page ou aplicação web (React, Vite, Next.js, Angular, Vue, Nuxt, Astro, HTML) para atingir nota máxima (98-100) no Google Lighthouse 13 em Performance, Accessibility, Best Practices, SEO e Agentic Browsing, aplicando técnicas avançadas de SEO, AEO (Answer Engine Optimization) e GEO (Generative Engine Optimization para ChatGPT, Perplexity, Claude e Gemini). Use quando o usuário pedir para melhorar SEO, AEO, GEO, ser encontrado por IAs/buscadores, criar llms.txt/ard.json ou bater nota 100 no Google Lighthouse.

Run one bounded refinement pass on a living Linear research ticket. Use for tickets structured with findings, open questions, a decision log, and a refinement protocol (e.g., DEV-47). Verify decaying claims against current sources, update findings in place, and log resolved decisions with evidence.

Live SEO data via DataForSEO MCP server: SERP analysis, keyword research (volume, difficulty, intent, trends), backlink profiles, on-page analysis, competitor and content analysis, business listings, AI visibility (LLM mention tracking), and domain analytics. Requires DataForSEO extension installed. Use when user says "dataforseo", "live SERP", "keyword volume", "backlink data", "AI visibility check", or "real search data".

Pick the right skill or skill chain for a task, load it in the right order, and ask only the questions that actually block the work. Works against any installed skill library of any size — it indexes whatever skill directories exist and needs no knowledge of their names. Use this on every request that involves real work — building or fixing code, designing or reviewing UI, debugging, refactoring, deploying, research, writing, pitch decks, documents, spreadsheets, slides, animation, writing, or auditing an existing project — and especially when the request is vague, spans several areas, is brand new, or you are not certain which installed skill applies. Also use it when the user says "route this", "which skill", "use whatever fits", "figure out the right approach", or hands over a task with no explicit tooling. Skip only for pure conversation, one-line factual questions, and follow-ups inside a workflow already routed.

Anti-slop art direction for native SwiftUI apps (iOS, iPadOS, watchOS, macOS, visionOS), the mobile port of leonxlnx/taste-skill. A script deals researched directions (idea, composition, type, palette, shape, motion) instead of letting the model pick the category default; a native-first ladder (system component, styled system component, custom component only when the idea needs it) keeps the result at home on the platform; a Swift linter and a Simulator capture matrix check the rules; a critique loop judges the screenshots. Use whenever you design, build, redesign, restyle, polish or review an app screen, flow, onboarding, home, paywall, settings, widget or component in SwiftUI, and whenever the user says an app looks generic, templated, "AI-made", default or asks for taste, premium, award-level, feature-worthy design.

Research a company before an interview and produce a sourced, dated one-page brief: what they sell, to whom, pricing and packaging, competitors and differentiation, recent launches and news, business health, and what all of it means for the product marketing role. Use before a recruiter screen, hiring manager, panel, or exec conversation.

Writes a competitor brief or sales battlecard: positioning, recent moves, SWOT, and how to win, with every claim sourced. Use whenever someone asks about a competitor, how two products compare, or for talking points against a rival, even with no web access and only a pasted page or notes to go on.

Audit version freshness, FR/EN parity, and metadata quality of all whitepapers and recap cards

Add descriptive prose to bare sections in whitepapers (FR + EN) using 9 parallel agents

Kill a business or product idea in one hour instead of three months, using three searches and four checks that catch the problems competitor research misses. Use this whenever someone is weighing up, pitching, comparing or "just thinking about" a business idea, app idea, side project or new product — including when they're already excited and only want help building it, and including when they ask "is this a good idea?", "has anyone done this?", "should I build X?" or describe a problem they want to solve for money.

Read an academic paper from a PDF, link, DOI, title, or text. By default, provide a one-sentence takeaway and field, standard analysis, then sequential paragraph translation with inline figure, equation, and table explanations, followed by research extensions.

Run an agile-style sprint retrospective on a completed Claude collaboration session. Use this skill after finishing a piece of work with Claude — a scope of work, a data analysis, a content build, a code session, a Cowork task — to reflect together on what worked, what didn't, and what to change. Triggers on: "let's do a retro", "run a retro", "retrospective", "retro-bot", "what worked well", "what could we improve", "session review", "what should we do differently", "let's debrief", or any request to reflect on a completed Claude session and extract improvements. The skill produces a collaborative retro board, maps issues to actionable file changes (skill updates, Claude.md edits, process changes), applies those changes using proper versioning, and saves a dated snapshot to a user-configured archive directory — building a permanent audit trail of every retro run and its outcomes. Always use this skill rather than improvising a retro.

查找和阅读用户本地 Zotero 文献。当用户提到 Zotero 文献、Zotero 中的论文、文献库、LLM-for-Zotero 或 MinerU 提取，并希望按文章名称查找、检查提取状态、阅读、总结或分析论文时使用。先确认已下载的 PDF，再核对 Markdown 解析；优先基于 Markdown 阅读。也可显式调用 Zotero-paper-reading。一般性 Zotero 软件安装或插件配置问题不触发本技能。

Create polished, offline, editable ArcGIS Pro thematic maps for reports, presentations, and spatial analysis at city, county, town, planning-zone, and village scales.

Create atomic Git commits using Conventional Commits. Use when the user asks to generate or create a commit; use quick mode by default and full mode only when complete diff analysis is explicitly requested.

>- Azure Boards / ADO cards (tickets, work items, stories, tasks). Use when the user mentions their board, sprint, plate, or an item like #512 — even without saying Azure DevOps. Branches: set up the ado skill; show my cards; create or break into tasks; move or close; log time; comments; description or acceptance criteria; plan my sprint / what first. Board operations only (not app code, SQL, or pipelines).