Agent Skills

---

name: ppt-animation

description: 生成 PPT 风格的翻页 HTML 演示动画，支持暗色炫酷、暖色报纸、简约白色等多套主题，适合视频科普、技术讲解场景

---

name: stacked-data-cards

description: 生成「叠放式数据论点卡」自动播放动画的单文件 HTML：白色论点卡从底部依次弹入居中，旧卡向左叠成牌堆，卡内数据组件（折线描边、数字滚动、环形仪表、对比条、数据表格、图标卡组）按毫秒级时间线级联入场，整段时长对齐口播/旁白秒数，录屏即成片。适合视频中的数据论证、研究结论展示、观点陈列、反直觉点揭示等场景。

---

name: swiss-shot-demos

description: 制作"瑞士国际主义风格（Swiss Style）+ 图表驱动"的视频分镜演示动画——纸灰底/炭黑墨/发丝线/唯一强调橙的编辑部级排版，每页用匹配内容的可视化图表（时间线/循环环/流程链/对照表/点阵计数/热力矩阵…）代替文字堆砌，一镜一 HTML，电影化播放器全屏录屏即成片。只要用户提到：瑞士风格演示页、Swiss Style 动画、图表驱动的分镜演示、把口播稿做成可视化动画、信息图风格演示动画、或要求把一批演示页从暗色油光改成瑞士风/杂志编辑风——就用本 Skill。内容完全开放不设限：技术教程、产品发布、安全科普、知识叙事皆可。与 video-shot-demos（风格轮换体系）同底盘，但本 Skill 全系列统一一套瑞士设计语言，靠图表类型与版式变化制造镜头差异。

---

name: ui2v-motions

description: UI2V 动效库——1,262 个 GSAP 动效包（616MB 全量离线镜像，源自 ui2v.com / illli-studio/motions）+ 按镜头用途的选型索引 + 接入演示动画的完整配方。给视频演示页/产品页/落地页挑选和嵌入高质量 UI 动效：图表入场、列表滚动、卡片翻转、文字编排、数字滚动、背景循环等即插即用。只要用户提到：UI 动效、动效库、ui2v、GSAP 动效素材、给演示页加动效、motion 素材、页面动效不够炫——就用本 Skill。与 swiss-shot-demos / video-shot-demos 等演示动画技能配套：动效包的 paused GSAP 时间轴可直接挂到 cue 时间轴上驱动。

---

name: video-shot-demos

description: 为视频/课程/解说文案制作"每镜头一个 HTML"的高完成度网页演示动画（分镜演示页、视频配套演示动画、录屏用动画网页、科普/技术/产品发布可视化镜头）。只要用户提到：给视频做演示动画网页、分镜 HTML、shot 页面、把口播稿/大纲做成动画、视频画面设计网页版、录一段网页动画、或要求改造升级一批已有演示页——就用本 Skill。核心是统一的电影化播放器机制 + 每页完全不同的视觉风格 + 镜头语言 + 同步音效 + 角色表情吐槽，内容完全开放不设限。

---

name: win11-ui-demos

description: 制作"仿 Windows 11 系统 UI"的写实仿真网页演示动画——观众像在看一台真 Win11 电脑的精心编排录屏：桌面、任务栏、资源管理器、开始菜单、右键菜单、系统对话框、通知、WinRE 蓝屏 1:1 还原，光标演出引导操作，教学标记层叠加讲解，电影化播放器全屏录屏即成片。只要用户提到：仿 Win11/Windows 演示、系统 UI 仿真动画、电脑教学演示页、软件操作教程动画、把 Windows 操作过程做成动画、或要求改造升级一批系统界面风格演示页——就用本 Skill。内容开放不设限：任何电脑操作、系统知识、软件教程、安全科普都能转译成"真电脑里发生的事"。

---

name: 学霸笔记

description: 生成手写笔记本风格的单文件 HTML 学习笔记。当用户需要将技术内容、漏洞分析、知识总结等转化为视觉精美的网页笔记时使用。

---

name: intelligent-prompt-generator

description: 智能提示词生成器 v2.0 - 支持人像/跨domain/设计三种模式，语义理解、常识推理、一致性检查

---

name: intelligent-prompt-generator

description: 智能提示词生成器 v2.0 - 支持人像/跨domain/设计三种模式，语义理解、常识推理、一致性检查

---

name: prompt-analyzer

description: 提示词分析与洞察 - 查看Prompt详情、对比差异、推荐相似提示词、元素库统计

---

name: prompt-analyzer

description: 提示词分析与洞察 - 查看Prompt详情、对比差异、推荐相似提示词、元素库统计

---

name: universal-learner

description: 通用学习器 - 从任何领域的Prompt中自动提取可复用元素，持续学习和积累知识

---

name: universal-learner

description: 通用学习器 - 从任何领域的Prompt中自动提取可复用元素，持续学习和积累知识

---

name: topology

---

name: build-review-interface

description: Build a custom browser-based annotation interface tailored to your data for reviewing LLM traces and collecting structured feedback. Use when you need to build an annotation tool, review traces, or collect human labels.

---

name: error-discovery

description: Run error analysis on a dataset. Build a review UI, select diverse samples, monitor annotations, and organize failure modes.

---

name: eval-audit

description: Audit an LLM eval pipeline and surface problems: missing error analysis, unvalidated judges, vanity metrics, etc. Use when inheriting an eval system, when unsure whether evals are trustworthy, or as a starting point when no eval infrastructure exists. Do NOT use when the goal is to build a new evaluator from scratch (use error-discovery, write-judge-prompt, or validate-evaluator instead).

---

name: evals-start

description: Entry point for evals. Use when the user asks for help with evals, does not know where to begin, or asks for something no other skill in this plugin matches. Do NOT use when a more specific skill in this plugin already matches; load that skill directly.

---

name: evaluate-rag

description: Guides evaluation of RAG pipeline retrieval and generation quality. Use when evaluating a retrieval-augmented generation system, measuring retrieval quality, assessing generation faithfulness or relevance, generating synthetic QA pairs for retrieval testing, or optimizing chunking strategies.

---

name: generate-synthetic-data

description: Create diverse synthetic test inputs for LLM pipeline evaluation using dimension-based tuple generation. Use when bootstrapping an eval dataset, when real user data is sparse, or when stress-testing specific failure hypotheses. Do NOT use when you already have 100+ representative real traces (use stratified sampling instead), or when the task is collecting production logs.

---

name: validate-evaluator

description: Calibrate an LLM judge against human labels using data splits, TPR/TNR, and bias correction. Use after writing a judge prompt (write-judge-prompt) when you need to verify alignment before trusting its outputs. Do NOT use for code-based evaluators (those are deterministic; test with unit tests per `write-code-eval`).

---

name: write-code-eval

description: Write code evaluators for known failure modes with objective rules. Use when code can check the rule from a trace, with or without a reference answer. Use `write-judge-prompt` when the rule requires interpretation.

---

name: write-judge-prompt

description: Design LLM-as-Judge evaluators for subjective criteria that code-based checks cannot handle. Use when a failure mode requires interpretation (tone, faithfulness, relevance, completeness). Do NOT use when the failure mode can be checked with code (regex, schema validation, execution tests); use `write-code-eval`. To validate an existing judge, use `validate-evaluator`.

---

name: arkon-edit

description: Propose or directly apply edits to Arkon wiki pages, including proposing brand new pages. Contributors create drafts for review; editors/admins can edit/create directly. Triggers on: update wiki, fix this page, propose edit, edit wiki page, correct the KB, improve wiki, resubmit my draft, withdraw my draft, create new wiki page, propose new page.