web-video-presentation

把一篇文章或口播稿，做成"看起来像视频"的点击驱动 16:9 网页演示，可选合成口播音频。流程：原始文章 → **一次产出**口播稿 + outline 开发计划 → 用户**一次对齐** 5 件事（稿子 / outline / 主题 / 素材 / 开发模式）→ 网页开发（逐章 / 顺序 / 并行）→ 可选音频合成（provider-agnostic：内置 MiniMax mmx-cli + OpenAI TTS，可换 ElevenLabs / edge-tts / Azure / 自带 TTS）。**outline 只规划节奏与信息密度，不规划动画** —— 动画由章节开发时按 PRINCIPLES + ANTI-AI 法则即时设计。每次点击推进口播稿的一个节拍，每一步独占整屏，进度条平时隐藏只在悬浮时出现。适用场景：用网页做视频（动态 PPT 但不像 PPT）、把口播稿 / 文章变成可交互的解说、为 B 站 / YouTube / 视频号录屏教程、做有电影感的产品 / talk demo。本 Skill 沉淀的是设计方法论 + 协作流程 —— 不绑定任何特定样式 / 字体 / 颜色 —— 因此能复用到任意主题与美学。

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

77%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with a clear, checkpointed workflow and strong self-check feedback loops. Its main gaps are some redundancy that hurts conciseness, and two referenced bundle paths (templates/scripts/tts-providers/, themes/) that do not exist in the bundle despite being relied on by scaffold.sh and the theme-selection checkpoint.

Suggestions

Ship the missing 'themes/' directory (with theme.json + tokens.css per theme) and 'templates/scripts/tts-providers/' (README.md, minimax.sh, openai.sh) that scaffold.sh copies from, so the referenced paths resolve.

Remove the 十条原则 duplicate table (or reduce it to a one-line pointer) since CHAPTER-CRAFT.md Part 0 already owns it; collapse repeated CHAPTER-CRAFT.md cross-references.

Move the full 总结模板 skeleton block into a reference file and keep only a short pointer in the body to tighten the token budget.

Dimension	Reasoning	Score
Conciseness	Mostly efficient operational guidance (tables, terse bullets, code blocks) but ~450 lines with redundancy — the 十条原则 table duplicates CHAPTER-CRAFT.md Part 0, CHAPTER-CRAFT.md is re-referenced many times, and the 总结模板 skeleton could be tightened; anchor 3 ('every token earns its place') is not met, so anchor 2 is closest.	2 / 3
Actionability	Provides copy-paste-ready commands — 'bash scripts/scaffold.sh ./presentation --theme=<id>', 'rm -rf presentation/src/chapters/01-example', 'npx tsc --noEmit', 'PRESENTATION_TTS=openai npm run synthesize-audio', '?auto=1' — plus concrete file paths and hard rules (per-chapter CSS prefix, bump STORAGE_KEY).	3 / 3
Workflow Clarity	Clear phased sequence (Phase 1→4) with explicit hard checkpoints (Checkpoint Plan, Checkpoint Audio, 第1章强制验收) and feedback loops (硬性自检协议: 自检→修复→再汇报, Agent Teams→subAgent→自检 escalation, tsc --noEmit gate) for batch/destructive-style chapter work.	3 / 3
Progressive Disclosure	Structure is well-signaled (one-level-deep references/, a '何时读' navigation table), but two referenced paths are absent from the actual bundle — 'templates/scripts/tts-providers/README.md' and the entire 'themes/' dir (themes/*/theme.json) that scaffold.sh --theme/--list-themes and Checkpoint Plan depend on — which per the bundle-structure guideline breaks navigation and prevents a 3.	2 / 3
	Total	10 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, action-oriented, and supplies an explicit use-when clause with natural trigger phrasing, covering what and when clearly. Its only weakness is density/length — it packs the entire workflow into the description rather than staying a tight summary.

Suggestions

Trim the embedded workflow narrative ('流程：原始文章 → 一次产出…') from the description; that detail belongs in the body, leaving the description as a concise what+when summary.

Lead with the trigger clause ('Use when…' / '适用场景') so the natural keywords appear before the process mechanics.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — '做成点击驱动 16:9 网页演示', '一次产出口播稿 + outline 开发计划', '音频合成（内置 MiniMax mmx-cli + OpenAI TTS）' — rather than vague language.	3 / 3
Completeness	Explicitly answers both what (article→web presentation with optional TTS) and when (an explicit '适用场景' / use-when clause listing trigger scenarios), so it is not capped at 2.	3 / 3
Trigger Term Quality	The '适用场景' clause surfaces natural user phrasing — '用网页做视频（动态 PPT）', '把口播稿 / 文章变成可交互的解说', '为 B 站 / YouTube / 视频号录屏教程' — that a user would actually say.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche — click-driven 16:9 web 'video' from articles/scripts with per-step full-screen scenes — with distinct triggers unlikely to collide with other skills.	3 / 3
	Total	12 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 2 missing, 1 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 2 missing, 1 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: ConardLi/garden-skills
Path: skills/web-video-presentation/SKILL.md
Commit: aaf9a82

Reviewed: ~1 month ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.