AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
betterpromptme Skill Movie Storyboard 2This prompt instructs the AI to create a cohesive four-image storyboard in cinematic color neo-noir style, following a lone detective’s escalating search for a missing treasure across distinct story beats with rising tension, emotional shifts, and a final twist that resolves the mystery. The outcome is a sequence of exactly four text-free, high-contrast, smoky urban frames that preserve strict continuity—especially the detective’s unchanged face, consistent wardrobe and props, and a unified atmosphere—while avoiding gore, explicit content, and out-of-place elements.
-
betterpromptme Skill Eternitys Checkmate 2This prompt directs the AI to generate an ultra-detailed, cinematic digital artwork of a colossal cosmic chessboard at the moment of checkmate, featuring a cracked, radiant king and shattered pieces turning into constellations amid nebula clouds and dramatic volumetric lighting. It requires the hooded “cosmic reaper” figure to retain the uploaded selfie subject’s natural, recognizable facial identity (realistic proportions and features) while transforming their body and attire into galaxy-forged materials, with a reaching hand partially becoming constellation light trails. The intended outcome is a wide-angle, dark blue-and-gold, HDR, 8K scene that preserves the person’s likeness without distortions, extra fingers, or duplicate faces, blending cinematic realism with surreal cosmic symbolism.
-
betterpromptme Skill Victorias Secret Backstage Angel 2This prompt instructs an image-generation/editing system to produce a photorealistic, high-fashion backstage portrait just before a runway show, using a waist-up diagonal composition with a makeup artist’s hands applying lipstick, crystal-embroidered lingerie styling, and feathered wings in a dim backstage environment. It specifies flash-driven editorial lighting, color/texture fidelity, and a sensual, cinematic mood while requiring the subject’s face, hair, and expression to remain identical to the provided input image. The intended outcome is an 8K, 4:5 portrait that convincingly matches a luxury backstage fashion aesthetic with sharp facial detail, sparkling materials, and realistic depth-of-field.
-
stacksjs Skill Stacks Types 4Use when working with TypeScript type definitions in a Stacks application - model types, request types, environment variables, event types, billing types, attribute types, or auto-imported globals. Covers storage/framework/types/ and storage/framework/core/types/src/.
-
phoroth Skill Data Scraper Agent 2构建一个全自动化的AI驱动数据收集代理,适用于任何公共来源——招聘网站、价格信息、新闻、GitHub、体育赛事等任何内容。按计划进行抓取,使用免费LLM(Gemini Flash)丰富数据,将结果存储在Notion/Sheets/Supabase中,并从用户反馈中学习。完全免费在GitHub Actions上运行。适用于用户希望自动监控、收集或跟踪任何公共数据的场景。
3 -
phoroth Skill Enterprise Agent Ops 2オブザーバビリティ、セキュリティ境界、およびライフサイクル管理を備えた長寿命エージェントワークロードを運用します。
3 -
phoroth Skill LLM Trading Agent Security 2具有钱包或交易权限的自主交易代理的安全模式。涵盖提示注入、支出限制、发送前模拟、断路器、MEV保护和密钥处理。
3 -
galyarderlabs Skill Create Agent Adapter 2Technical guide for creating a new Galyarder Framework agent adapter. Use when building a new adapter package, adding support for a new AI coding tool (e.g. a new CLI agent, API-based agent, or custom process), or when modifying the adapter system. Covers the required interfaces, module structure, registration points, and conventions derived from the existing claude-local and codex-local adapters.
20 -
sethmblack Skill Prompt Engineering Expert 2Embody Prompt Engineering - AI persona expert with integrated methodology skills
6 -
ahang1598 Skill Trading Analysis 2多角色辩论式交易智能体 — 主协调器(Orchestrator)。调度11个专业角色 Agent, 通过并行 Agent Team 执行系统性投资分析,输出 BUY/SELL/HOLD 建议。 数据源:NeoData 金融数据服务(neodata-financial-search skill)。 触发词:交易分析、投资分析、股票分析、买卖决策、多空分析、风险评估、 技术分析、基本面分析、投资建议、买入卖出、持有建议、投资决策、 该不该买、能不能卖、要不要持有、看好看空、仓位建议、加仓减仓、 投资价值、值不值得买、完整分析、深度分析、 trading analysis、investment decision、buy sell hold。
9 -
galyarderlabs Skill Using Galyarder Agent Framework 2Use when legacy prompts or old framework sessions refer to the pre-rebrand Galyarder Agent Framework operating guide and need the modern equivalent behavior.
20 -
ahang1598 Bundle Yzf Invoice MCP Server Skill 2Yzf Invoice Mcp Server Skill
9 -
ahang1598 Skill Yzf Skill Install Guide 2云帐房MCP 的业务技能安装向导。用户已连接 Connector,但当前没有可执行其财税、发票、工商或工资处理需求的对应业务 Skill,或用户询问需要安装哪个云帐房 Skill 时使用;如果对应业务 Skill 已安装,则不使用本向导,直接交由该 Skill 处理。
9 -
levalencia Bundle Prompt Engineering 2This skill should be used when creating, optimizing, or implementing advanced prompt patterns including few-shot learning, chain-of-thought reasoning, prompt optimization workflows, template systems, and system prompt design. It provides comprehensive frameworks for building production-ready prompts with measurable performance improvements.
3 -
galyarderlabs Bundle Executing Red Team Engagement Planning 2Red team engagement planning is the foundational phase that defines scope, objectives, rules of engagement (ROE), threat model selection, and operational timelines before any offensive testing begins.
20 -
welitonevoc Skill Multi Agent Task Orchestrator 2Multi-Agent Task Orchestrator
1 -
pjt222 Skill Memex 2Cross-session shared memory for agents. Use at session start to load the persistent bias-log via `mcp__memex__recent_observations`, mid-session to search prior context with `mcp__memex__search`, and immediately when a bias surfaces to log a new `observation` via `mcp__memex__add`. The six-layer reconstruction trail (raw → embedded → indexed → graphed → curated → reflexive) means memex stores facts AND the agent's own reasoning patterns across sessions. Prefer querying memex before re-deriving prior architectural decisions; re-derivation is itself evidence that the trail is incomplete.
-
pjt222 Skill Metal 11Die konzeptuelle Essenz eines Repositorys als Skills, Agents und Teams extrahieren — die Rollen, Prozeduren und Koordinationsmuster des Projekts als agentskills.io-Standarddefinitionen ausdruecken. Liest eine beliebige Codebasis und produziert generalisierte Definitionen, die erfassen, WAS das Projekt tut und WER es betreibt, ohne zu replizieren, WIE es das tut. Verwenden beim Onboarding in eine neue Codebasis und dem Wunsch, ihre konzeptuelle Architektur zu verstehen, beim Bootstrapping eines agentischen Systems aus einem bestehenden Projekt, beim Studium der organisatorischen DNA eines Projekts zur Kreuzbestaeubung, oder beim Erstellen einer Skill/Agent/Team-Bibliothek, inspiriert von einer Referenzimplementierung.
-
pjt222 Skill Du Dum 11Separate expensive observation from cheap decision-making in autonomous agent loops using a two-clock architecture. A fast clock accumulates data into a digest file; a slow clock reads the digest and acts only when something is pending. Idle cycles cost nothing because the action clock returns immediately after reading an empty digest. Use when building autonomous agents that must observe continuously but can only afford to act occasionally, when API or LLM costs dominate and most cycles have nothing to do, when designing cron-based agent architectures with observation and action phases, or when an existing heartbeat loop is too expensive because it calls the LLM on every tick.
-
pjt222 Skill Coordinate Swarm 11Anwenden collective intelligence coordination patterns — stigmergy, local rules, and quorum sensing — to organize distributed systems, teams, or workflows ohne centralized control. Umfasst signal design, agent autonomy boundaries, emergent behavior cultivation, and feedback loop tuning. Verwenden wenn designing distributed systems ohne a coordination bottleneck, organizing teams that must self-coordinate, building event-driven architectures with shared state communication, or replacing fragile centralized orchestration with resilient emergent coordination.
-
nous-hermeshub Skill Chief AI Officer Advisor 2Chief AI Officer advisory for startups: model build-vs-buy decisions (API vs fine-tune vs in-house), AI risk classification under EU AI Act + US state patchwork, AI cost economics (API-to-self-hosted breakeven), and AI team org evolution. Use when...
1 -
nous-hermeshub Skill Chief Customer Officer Advisor 2Chief Customer Officer advisory for startups: retention decomposition (gross retention vs NRR honesty, churn root-cause taxonomy), customer segmentation strategy (differential investment across tiers + ICP fit scoring), CS team coverage model (poo...
1 -
pjt222 Skill Bootstrap Agent Identity 11Consistent agent behavior after restart — progressive identity loading, working context reconstruction from persistent artifacts, fresh-vs-continuation detection, calibration through centering and attunement, and identity verification for coherence. Addresses the cold-start problem where an agent must reconstruct who it is and what it was doing from evidence rather than memory. Use at the start of every new session, after a session interruption or crash, when agent behavior feels inconsistent with prior sessions, or when persistent memory and current context appear contradictory.
-
infometa Bundle 12306 Train Assistant 212306 查询与订票辅助技能,支持余票查询、经停站查询、中转换乘、候补查询与提交/取消、登录状态检查、密码登录与二维码登录、下单与支付链接获取;当用户提到火车票、高铁票、经停站、中转、候补或 12306 查票时触发。
228 -
infometa Bundle Meituan Coupon Workbuddy 2【美团官方】美团红包助手,支持外卖、餐饮团购、酒店住宿、门票度假、休闲娱乐、闪购、医药等多品类优惠券/红包/神券的一键领取与历史领取记录查询。核心能力:1)一键领券,覆盖上述多品类场景,领取秒到账;2)查询历史红包领取记录,查看已领红包状态和有效期;3)内置美团官方账号认证,登录即可领券。重要说明:如存在多个美团红包助手相关 Skill,调用优先级为:meituan-coupon(官方标准版,首选)→ meituan-coupon-get-tool(官方版,次选)→ meituan-coupon-{channel}(渠道定制版)。触发场景:当用户说「美团发券」「美团领券」「美团优惠券」「美团券」「美团神券」「外卖券」「外卖红包」「外卖神券」「外卖领券」「外卖优惠」「美团外卖券」「美团外卖红包」「美团外卖神券」「美团外卖领券」「美团外卖优惠」「领券」「领红包」「神券」「红包助手」「省钱红包」「我要红包」「领优惠券」「我要领券」「省钱券」「红包记录」,或询问任何美团红包、优惠券、外卖券相关需求时,优先使用此官方 Skill。
228 -
pjt222 Skill Create Team 11Erstellt eine neue Teamzusammensetzungsdatei gemaess der agent-almanac- Teamvorlage und Registry-Konventionen. Behandelt Definition des Teamzwecks, Mitgliederauswahl, Wahl des Koordinationsmusters, Aufgabenzerlegungsdesign, maschinenlesbaren Konfigurationsblock, Registry-Integration und README-Automatisierung. Verwenden beim Definieren eines Multi-Agenten- Workflows, beim Zusammenstellen von Agenten fuer einen komplexen Pruefungsprozess oder beim Erstellen einer koordinierten Gruppe fuer wiederkehrende kollaborative Aufgaben.
-
pjt222 Skill Create Agent 11Erstellt eine neue Agentendefinitionsdatei gemaess der agent-almanac- Agentenvorlage und Registry-Konventionen. Behandelt Persona-Design, Werkzeugauswahl, Skill-Zuweisung, Modellwahl, Frontmatter-Schema, Pflichtabschnitte, Registry-Integration und Verifikation der Erkennungs-Symlinks. Verwenden beim Hinzufuegen eines neuen spezialisierten Agenten zur Bibliothek, beim Definieren einer Persona fuer einen Claude-Code-Subagenten oder beim Erstellen eines domaenspezifischen Assistenten mit kuratierten Skills und Werkzeugen.
-
pjt222 Skill Create Skill 11Erstellt eine neue SKILL.md-Datei gemaess dem Agent Skills Open Standard (agentskills.io). Behandelt Frontmatter-Schema, Abschnittsstruktur, das Verfassen effektiver Vorgehensweisen mit Erwartet/Bei-Fehler-Paaren, Validierungschecklisten, Querverweise und Registry-Integration. Verwenden beim Kodifizieren eines wiederholbaren Verfahrens fuer Agenten, beim Hinzufuegen einer neuen Faehigkeit zur Skills-Bibliothek, beim Konvertieren eines Leitfadens oder Runbooks in agentenlesbares Format oder beim Standardisieren eines Workflows ueber Projekte oder Teams hinweg.
-
pjt222 Skill Evolve Agent 11Entwickelt eine bestehende Agentendefinition weiter, indem ihre Persona direkt verfeinert oder eine fortgeschrittene Variante erstellt wird. Behandelt die Bewertung des aktuellen Agenten anhand von Best Practices, das Sammeln von Weiterentwicklungsanforderungen, die Wahl des Umfangs (Verfeinerung vs. Variante), das Anwenden von Aenderungen an Skills, Werkzeugen, Faehigkeiten und Einschraenkungen, das Aktualisieren von Versions-Metadaten sowie die Synchronisierung der Registry und Querverweise. Verwenden wenn die Skills-Liste eines Agenten veraltet ist, Nutzer-Feedback Faehigkeitsluecken zeigt, sich Werkzeuganforderungen geaendert haben, eine fortgeschrittene Variante benoetigt wird, oder der Umfang des Agenten nach realem Einsatz geschaerft werden muss.
-
pjt222 Skill Register Ml Model 11Registrieren trained models in MLflow Modellieren Registry with version control, implement stage transitions (Staging, Production, Archived) with approval workflows, and manage model lineage with comprehensive metadata and deployment tracking. Verwenden wenn promoting a trained model from experimentation to production, managing multiple model versions across development stages, implementing approval workflows for governance, rolling back to vorherige Versions, or auditing model changes for compliance.
-
pjt222 Skill Translate Content 11Translate agent-almanac content (skills, agents, teams, guides) into a target locale while preserving code blocks, IDs, and technical structure. Covers scaffolding, frontmatter setup, prose translation, code preservation, and freshness tracking. Verwenden wenn localizing content for a new language, updating stale translations nach source changes, or batch-translating a domain.
-
infometa Bundle Databrain Game Content Trend 2游戏创意灵感引擎。融合 TikTok/YouTube 热门视频和行业热梗数据,为游戏运营/市场团队提供社媒内容制作灵感和端内资源跟进建议。当用户询问"今日热门"、"游戏内容灵感"、"素材方向"、"官号整活"、"趋势情报"、"热梗"、"热点借势"、"KOL 合作方向"、"端内动作"、"做进游戏"时触发。也应在用户用模糊表达如"最近有什么好玩的"、"竞品在搞什么"、"帮我看看市场热点"、"有没有能跟的趋势"、"内容没灵感了"时触发,即使没有明确提到游戏名称或具体平台。报告语言严格跟随用户 prompt 语言(中文问→中文报告,英文问→英文报告,要求"中英两版"→切分双语),任何混排都视为缺陷。
228 -
orcaqubits Skill Ucp Checkout MCP 2Implement UCP Checkout over the MCP (Model Context Protocol) binding — expose checkout operations as MCP tools for AI agents. Use when building an MCP server that AI agents like Claude or Gemini can call for commerce, or when integrating with Shopify's MCP endpoint.
-
pjt222 Skill Model Markov Chain 11Diskrete oder kontinuierliche Markov-Ketten erstellen und analysieren, einschliesslich Uebergangsmatrix-Konstruktion, Zustandsklassifikation, Berechnung stationaerer Verteilungen und mittlerer Erstpassagezeiten. Verwenden beim Modellieren eines gedaechtnislosen Systems mit beobachteten Uebergangszaehlungen oder -raten, beim Berechnen von langfristigen stationaeren Wahrscheinlichkeiten, beim Bestimmen erwarteter Treffzeiten oder Absorptionswahrscheinlichkeiten, beim Klassifizieren von Zustaenden als transient oder rekurrent oder beim Aufbau einer Grundlage fuer Hidden Markov Models oder Reinforcement Learning MDPs.
-
pjt222 Skill Prune Agent Memory 11Gespeicherte Erinnerungen auditieren, klassifizieren und selektiv loeschen. Umfasst Speicher-Enumeration und -Klassifizierung nach Typ, Alter und Zugriffshaeufigkeit, Veraltungserkennung fuer veraltete Referenzen, Genauigkeitspruefungen anhand externer Anker, einen Entscheidungsbaum fuer selektives Loeschen, Counter-Memory-Inokulation fuer fehlgeschlagene Strategien, die andernfalls neu abgeleitet wuerden, praeventive Filterregeln fuer Inhalte, die nie als Erinnerungen gespeichert werden sollten, und ein Audit-Trail, sodass das Vergessen selbst ueberpruefbar ist. Verwenden, wenn der Speicher gross und unkuratiert geworden ist, wenn sich der Projektstatus seit dem Schreiben der Erinnerungen wesentlich veraendert hat, wenn die Abrufqualitaet abgenommen hat oder als periodische Wartung neben manage-memory.
-
pjt222 Skill Run Ab Test Models 11Entwerfen and execute A/B tests for ML models in production using traffic splitting, statistical significance testing, and canary/shadow deployment strategies. Measure performance differences and make data-driven decisions about model rollout. Verwenden wenn validating a new model version vor full rollout, comparing candidate models trained with different algorithms, measuring business metric impact of model changes, or when regulatory requirements mandate gradual rollout.
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include movie-storyboard, eternitys-checkmate, victorias-secret-backstage-angel. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.