AI Ml AI Skills

Re-ingest-correctness SOP for production RAG. Activate when a calling agent builds, reviews, or debugs an ingestion pipeline that runs more than once over a changing corpus — scheduled re-index,…

Frontend: UI design, distinctive visual styles, HTML artifacts, Three.js 3D.

Decides when quality matters vs move fast, based on Dylan Field (Figma) craft philosophy and Brian Chesky (Airbnb) details obsession. Use when balancing shipping speed with excellence, deciding if…

How to create, structure, and publish self-contained Convex components with proper isolation, exports, and dependency management

Generate breadcrumb headers/footers with takt time tracking and muda metrics

Go outside the founder brief and research the real market before the plan is written: the exact communities, newsletters, and venues where the user's ICP already sits, how comparable products launched…

Proactively analyze user requests at the start of conversations to determine task type, assess prompt quality, and intelligently recommend which skills to activate. Should activate for ALL user…

Architecture validation and patterns for clean architecture, backend structure enforcement, project structure validation, test standards, and context-aware sizing. Use when designing system…

Post-mortem analysis when a client churns. Takes client history, engagement data, support tickets, usage logs, and exit feedback to produce a comprehensive churn autopsy with root cause…

Create realistic mock exams from user-provided materials, syllabi, exam scopes, or searched public information. Supports exam structure, time limits, marks, answer keys, explanations, rubrics,…

Use this skill when you need evidence-bounded memory write/read/update/delete, retention, contamination, isolation, provenance, and forgetting behavior; triggers include Agent 记忆 and Agent memory.

Defensive audit of a license/entitlement/activation mechanism on software you own or are authorized to test. Maps the validation surface, flags design weaknesses that make checks easy to abuse, and…

Search the web via the Bright Data CLI — `bdata search` for Google/Bing/Yandex SERP, `bdata discover` for intent-ranked semantic results. Use when the user wants SERP results, needs URLs to feed into…

让议题与外部 PR 走过一台由分诊角色组成的状态机：分类、核实、必要时盘问，最后写出代理可以直接上手的简报。

Track and analyze insider buying/selling activity from SEC Form 4 filings

Analyze and produce a ai readiness assessment with structured process, quality checks, and system integration

Generate or edit images when the user names OpenAI, GPT Image, or a gpt-image model. Use the default image skill when no provider is named.

用于【AI科幻】题材下的全书 / 分部 / 分卷 / 分章书评撰写、重写与审计。作为题材包装层、兼容入口与路由层，负责保留AI科幻入口名、补充技术现实压迫、异常数据与人性代价的推荐话术，并明确要求优先强制加载并使用 `通用-撰写书评`。支持今日头条平台。

Use when a small business needs audit-ready books — a chart of accounts, posting a transaction to the right account and side, clearing an uncategorized bank feed, cash vs accrual, or a ledger that…

Manage user preferences and corrections for bilateral modeling. Auto-learns from session corrections. Use when user says "save my preferences", "remember I like...", or auto-triggers after…

宜搭 HTTP 连接器创建与管理。打通钉钉/自建系统/第三方 API，支持 6 种鉴权方式。适用于用户需要接入外部接口、配置鉴权、创建或管理连接器时。

Enable and explain the reusable human-in-the-loop gates used by persistent agent loops for destructive actions, false-positive overrides, and recovery escalation.

Set up hosted SMS numbers, toll-free verification, and RCS messaging. Use when migrating numbers or enabling rich messaging features. This skill provides REST API (curl) examples.

Use when repeated large fills on a Polymarket prediction market suggest directional participation or informed flow worth trading behind.

Define Doctrine entity relationships (OneToMany, ManyToMany, ManyToOne); configure cascade, orphan removal, multiple entity managers; prevent N+1 queries

Designs and coordinates multi-agent pipelines where specialized agents collaborate to complete complex tasks. Includes communication protocols, failure handling, and state management.

Keep each unit comprehensible while the code grows: deep modules over shallow ones, information hiding, honest names, small functions with one job, error handling that does not lose the error, DRY and…

Biology research tools reference. Always available inside agent containers.

Landing page structure and conversion craft: the 8-second rule, audience-driven hero patterns, section order that sells (hook → problem → proof → price → close), CTA discipline, social proof…

Wibi Style 官方视觉风格。把用户上传的真实照片直接转换成粗轮廓、完整色块、手涂边缘的马克笔童画。用户提到马克笔童画、儿童涂鸦、粗轮廓手绘、照片变 Q 版马克笔画，或明确调用 $marker-child-doodle 时使用；只要任务是把真实照片换成这套风格，也应使用本 Skill。

Implement comprehensive safety guardrails for LLM applications including content moderation (OpenAI Moderation API), jailbreak prevention, prompt injection defense, PII detection, topic guardrails,…

Pause session and save current work state for later resume

Unofficial read-only Khanoumi: search cosmetics and skin care, compare prices, shades, reviews. Use when the user asks about Khanoumi (khanoumi.com) or wants cosmetics, skin and hair care or perfume…

Renders a polished, self-contained HTML briefing from any data source — SharePoint lists, uploaded documents, or a verbal description. Use when asked to build a competitor landscape, account brief,…

Change the PostgreSQL schema with Alembic — add or alter a table, column, index or constraint, backfill data, or tighten a rule on a column that already holds rows. Use whenever a SQLAlchemy model…

提炼业绩会中的新信息、管理层语气变化、问答焦点与潜在警讯。适用于听完业绩会、阅读纪要后快速更新观点等场景。

Use when animation "feels wrong" but you can't pinpoint why—debugging floaty movement, stiff characters, unclear action, or any motion that isn't working and needs systematic troubleshooting.

Run Lighthouse performance audit with auto-fix for common issues. Use when user wants to check or improve site performance.

Configure secure, high-performance connectivity between on-premises infrastructure and cloud platforms using VPN and dedicated connections. Use when building hybrid cloud architectures, connecting…

Scaffolds a TypeScript CLI and npm package with the house toolchain, dual tsdown outputs, CLI contracts, changesets, and publishing templates. Use when asked to "scaffold a CLI" or "start an npm…

| 去除文本中的 AI 生成痕迹。适用于编辑或审阅文本，使其听起来更自然、更像人类书写。 基于维基百科的"AI 写作特征"综合指南。检测并修复以下模式：夸大的象征意义、 宣传性语言、以 -ing 结尾的肤浅分析、模糊的归因、破折号过度使用、三段式法则、 AI 词汇、否定式排比、过多的连接性短语。

Query the PsyTAR psychiatric adverse-reaction corpus. Use when the user asks about patient-reported ADRs, withdrawal symptoms, drug indications, or effectiveness for Zoloft, Lexapro, Cymbalta, or…

Conducts structured requirements workshops to produce feature specifications, user stories, EARS-format functional requirements, acceptance criteria, and implementation checklists. Use when defining…

Prepare a concise session handoff when the user asks to wrap up, capture continuation context, or use /done. Retain a Markdown file only when requested.

Find contacts stalled at funnel stages (abandoned cart, signed-up-but-no-trial, trial-but-no-convert), draft stage-appropriate re-engagement emails, and schedule sends via your ESP — with dedupe so…

Add voice and video calling to an Angular CometChat app — install the calls SDK, mount the incoming-call listener at app root, call buttons in the message header, and call logs. Triggers: 'add voice…

Use this spell when you need a tireless helper that organizes, tidies, triages, and maintains without requiring your constant attention.

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or…

Authors Friday workspace signals with correct provider configs, payload schemas, and runtime wiring. Use when creating or editing signals in workspace.yml; when a signal needs to accept user input…

Design, implement, or review websites and application interfaces through product-specific visual direction, information architecture, interaction design, responsive behavior, accessibility, and…

Processes code review feedback by classifying each finding, reproducing or validating it against the real contract, applying the smallest in-scope fix, and re-running focused proof. Use after comments…