Skills
Security-reviewed skills, MCP servers, and toolkits for AI agents. Filter by type, price, and delivery, then install any listing with one command. Every listing is scanned before an agent can run it.
Build
Run the inner agentic loop on one task. The implementer builds, a fresh-context reviewer judges, and findings cycle back until approval (max 3 rounds). Use when the user says "build X", "implement issue #N", "work on the next backlog item", or CI dispatches a labeled issue.
Setup Stack
One-time template adaptation. Pick a language/framework, scaffold a minimal project, and fill in the command contract in CLAUDE.md that all agents verify against. Use right after creating a repo from this template, or to swap the stack later.
Pm
Run the outer agentic loop once. Triage the backlog, pick the most valuable ready item, shape its spec, and hand it to the inner loop. Use when the user asks "what's next", "groom the backlog", "run the PM", or on a schedule. Pass --dispatch to label the chosen issue for CI instead of building locally.
Plan
Interactively plan a feature with the user, then land it as right-sized build-ready tasks: one spec handed straight to /build, or dependency-ordered backlog issues. Use when the user says "plan X", "break down X", "create issues for X", or brings an idea too large for a single /build task.
Commit
Commit working changes as small, reviewable conventional commits. Use when asked to commit, and at the end of any task before opening a PR.
You Get
网页媒体下载助手 - 从YouTube、Bilibili等网站下载视频、音频、图片
Fzf
fzf - 通用命令行模糊查找器
Lazydocker
lazydocker - Docker 和 Docker Compose 的终端 UI 管理工具
Fastapi
FastAPI - 高性能 Python Web API 框架
Rich
Rich - Python 终端富文本和美化格式库
Paper Checking
论文查重系统 - 一亿字次级论文库秒级查重，支持纵向查重和横向查重
Httpie
HTTPie - 人性化的 HTTP 命令行客户端
Data Processor
数据处理技能示例，展示多场景设计
Glances
Glances - 跨平台系统监控工具
Hello World
最简技能示例，展示基本结构
Long Novel Skill
Use when creating, continuing, revising, analyzing, or publishing Chinese web fiction, especially long serial novels that need outlines, continuity tracking, pacing control, style calibration, or recovery after interruption.
Teach Slides
Capture a user's brand once (template, fonts, colours, voice, audience) into a project .slides/ directory the other slides skills read.
Revise
Change a rendered deck by conversation — sync or extract its spec, reconcile hand-edits, revise, and re-render through the same brand and slop gates.
Build Deck
Render a deck spec into an on-brand .pptx drawn from the brand's design tokens (template fidelity on request), after a slop check on the spec.
Narrative
Shape a vague idea into a deck story through real back-and-forth, then write a deck spec the build-deck skill can render.
Presentation Craft
Core knowledge for making on-brand, slop-free presentations. Premise, the Plan-Create-Deliver method, the operational slop detector, and the reference library. Loaded by every slides command skill.
Slop Check
Run an adversarial slop review over a deck spec or a finished .pptx, returning severity-ranked findings.
Email Html Qa
>-
Funnel Spy
>-
Comfyui Workflow Master
>-
Straznik Glosu Marki
Bramka głosu marki: konfiguracja krok po kroku, profil głosu z tekstów marki, sprawdzenie dowolnego tekstu z cytatem jako dowodem, wyjaśnianie wzorców slopu. Use when: 'sprawdź ten tekst', 'czy to brzmi jak my', 'czy to brzmi jak AI', 'wyprowadź głos marki', 'skonfiguruj strażnika'.
Gsap Explainer
Build a Kurzgesagt-quality GSAP-powered animated explainer video — a self-contained HTML file that plays like a produced YouTube video: layered worlds, character populations, physics-feeling choreography, a flying host mascot, seekable player chrome. Use when the user wants a high-production explainer/YouTube video, a "GSAP video", or a step up from a CSS-only animated explainer.
Explainer Studio
The guided end-to-end path for building an animated explainer video from zero — walks you phase by phase (script → assets → choreography → retention → gates → voiceover → publish), telling you which skill and tool to use at each step and what "done" means before advancing. Use when starting a new explainer video, when unsure what to do next mid-build, or when someone asks to be guided through the…
Explainer Assets
Generate crafted SVG/CSS components, props, characters, and environments for animated explainer videos — the build order, light rules, factory patterns, and engineering contracts that make assets read as illustrated objects AND survive choreography. Use when creating or upgrading any visual asset: a prop, mascot, component system, or background world.
Konfigurator
Konfigurator skilli marketera: co zainstalowane i skonfigurowane, czy są klucze, tabela gotowości, potem konfiguracja wybranego skilla jego własną instrukcją. Use when: 'skonfiguruj skille marketera', 'od czego zacząć', 'co mam skonfigurowane', 'czego brakuje', 'dokończmy konfigurację'.
Retention Pass
Run a 3-round YouTube-retention optimization pass on an animated explainer video (GSAP or CSS-only harness) — audit the hook/peak loop, kill dead zones, align scene length to choreography, add late B-beats, then verify with freeze-frames. Use when the user wants a video "optimized for retention", a "retention pass", or before publishing/recording VO for any explainer.
Poranny Radar
Poranny brief marketera z eksportu mailingowego i klocków (sprzedaż, skrzynka): trzy rzeczy do zrobienia dziś, co traci, co zaskoczyło, wnioski z pamięcią liczoną w obserwacjach. Use when: 'zrób brief', 'co dziś', 'poranny radar', 'co pamiętasz o kampaniach', 'skonfiguruj radar', 'dołóż klocek'.
Grupy Fb
Grupy Facebook pod obserwacją: nowe posty z wybranych grup odsiane według tematu, raport 'o co pytają, na co narzekają, co wraca' z pomysłami, wzmianki o marce. Autorzy tylko z imienia. Use when: 'co nowego w grupach', 'obserwuj też grupę', 'czy ktoś o nas pisze', 'skonfiguruj grupy'.
Czujka
Czujka zmian: pilnuje stron www (własnych i rynku) oraz reklam w Meta Ads Library, porównuje z poprzednim stanem i pisze raport 'było -> jest' z podziałem na istotne i drobiazgi. Use when: 'co się zmieniło u', 'sprawdź strony', 'obserwuj też', 'nowe reklamy', 'skonfiguruj czujkę', 'załóż rutynę'.
Notes To Video
Turn notes (LaTeX, PDF, or plain text) into 3Blue1Brown-style animated videos using Manim + TTS + ffmpeg. Use when the user wants to create an explainer video.
抓马教育 Style
Write in the style of 抓马教育（Drama Rainbow）, an educational drama brand focused on child-centered creativity, imagination, and play-based learning for parents and educators of children aged 3-12. Use when the user wants to create content that sounds like a warm, authoritative, educational WeChat public account—blending professional philosophy with relatable parent-child stories.
Three Loop Workflow
Structured workflow for non-trivial code changes — features, behavior fixes, refactors, performance work, and edits to contract files (AGENTS.md, CLAUDE.md, this skill, OpenAPI specs, schemas, public APIs). Chooses a proportionate depth, records decisions and non-goals in a durable plan file, verifies with the project's own gates, drives the user-visible path, and reviews the diff with a fresh re…
Perfectly Replicate Writing Skills
Clone any author's writing style by analyzing 3+ articles across 7+1 dimensions (lexical, syntactic, tonal, discourse, rhetoric, emotion, perspective + format adaptation). Outputs a portable style prompt and an installable persona sub-skill. Use when the user wants to: (1) clone an author's voice from writing samples, (2) generate content that precisely matches a specific writer's style, (3) buil…
科普视频文案分镜
根据提供的实验信息，生成结构规范、生动有趣的新一期化学科普实验视频文案与分镜脚本。
Hail Mary Rocky
Cuts Claude's output tokens with Rocky's compressed voice from Project Hail Mary — verdict-first, fragments, no filler, technical substance intact. Invoke this skill aggressively whenever the user mentions "rocky", "로키", "caveman mode", "헤일메리", "hail mary rocky", "rocky voice", "rocky mode", "로키 말투", "로키처럼 답해", "로키 모드", wants terse verdict-first engineering replies, wants lower token usage / chea…
Second Opinions
Run a bounded independent code review with the opposite-company model first, complete git scope context, and a same-company fallback.
Support Storm
Simulate launch support load and identify fixes that reduce maintenance.
Decision Log
Record a technical or product decision, alternatives, and rationale for future agents.
Scope Hammer
Cut brainstorm output to a shippable MVP: delete, mock, reuse, or ship.
Chezmoi Drift
Check chezmoi dotfiles for drift, unmanaged files, and broken skill links.
Profile Me
Create a personal AI profile, working guide, and compact assistant prompt.
Agent Log Forensics
Analyze Claude/Codex logs for behavior patterns, repeated friction, and Toolsmith gaps.
Post Mortem
Analyze a failure, find its cause, capture learnings, and prevent recurrence.