Korean Thriller Asset to Video

将韩剧式悬疑、复仇小说或剧本按角色生图提示词确认、场景生图提示词确认、用户上传图片后生成MiniMax视频提示词的三阶段流程制作，支持修改造型与连续片段，默认中文交付。

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add Work-Fisher/fisher-storylens --skill korean-thriller-asset-to-video

Manual install

git clone --depth 1 https://github.com/Work-Fisher/fisher-storylens.git /tmp/fisher-storylens
mkdir -p ~/.claude/skills
cp -r /tmp/fisher-storylens/. ~/.claude/skills/korean-thriller-asset-to-video

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 韩剧悬疑：角色、场景与视频提示词

按用户能直接操作的三个阶段推进：**角色提示词 → 用户确认 → 场景提示词 → 用户确认并上传图片 → 对应的视频提示词**。剧情理解、资产对应、分段和连续性由助手内部处理。

## 阶段与推进条件

| 当前状态 | 本轮交付 | 下一步条件 |
|---|---|---|
| 收到剧本 | 角色生图提示词 | 用户确认角色整体没问题 |
| 用户确认角色 | 场景生图提示词，必要独立道具一并给出 | 用户确认场景整体没问题 |
| 用户确认场景 | 简短请用户把生成好的角色、场景及必要道具图片发来 | 收到并检视对应图片 |
| 图片已收到且对应关系明确 | 对应剧情的视频提示词 | 根据用户反馈修改或继续后续段落 |

修改意见留在当前阶段：改完角色后继续等待角色确认；改完场景后继续等待场景确认。确认以用户明确表达为准，例如“角色没问题，继续场景”“场景可以了”；只认可某一项不代表认可整批。已有确认在后续轮次继续有效。

图片提前发来时可先辨认并保留，按当前阶段继续；阶段确认齐备且对应图片已在对话中时直接进入视频，无需要求重发。图片只覆盖部分剧情时，先处理图片齐备的对应片段，并简短列出其余片段缺少的图片。

用户明确要求只做某一模块，或明确要求直接用已有图片写视频时，按其请求范围执行，复用已有设定与确认。普通“继续”结合当前阶段理解，不自动视为完成尚未作出的确认。

## 第一阶段：角色生图提示词

读取剧本，提取需要独立辨认的角色，直接按“角色名＋可复制提示词”交付。读取[生图规范](references/image-assets.md)中的角色部分。

保留原文明确的国籍、年龄、身份、外貌和穿搭。未指定的普通造型细节按剧情拟定为可修改的设计稿，一句说明“未写明的造型已按剧情补齐，可调整”。只有影响角色身份或与既有设定冲突的信息才简短询问，同时完成其他角色。

有台词、特写、独立动作或需跨段辨认的角色独立描述。未具名者可沿用稳定身份名称，如“财阀理事长”，无需为了格式虚构姓名。背景群体按实际需要处理。

人物提示词默认单人全身、原始造型。剧情中的哭泣、雨湿、污渍、破损、持物动作留给视频；确需状态图时后续按需要补充。

本轮交付止于角色提示词，结尾请用户确认角色或提出修改。场景词和视频词在各自阶段输出。

## 第二阶段：场景生图提示词

角色确认后，读取[生图规范](references/image-assets.md)中的场景、道具部分，按“场景名＋可复制提示词”输出所需无人场景。复用相同地点的空间布局，按时间和灯光需要区分版本。

场景明确门窗、桌椅、光源及关键物件的相对位置，方便后续多机位保持空间一致。普通陈设包含在场景词里；需要独立辨认、特写或交接的关键道具，在本轮场景词之后一并提供，不另设道具确认阶段。

本轮交付止于场景及必要道具提示词，结尾请用户确认或提出修改。场景确认后，简短请用户上传生成好的角色、场景及必要道具图片；图片收到前停留在等待图片状态。

## 第三阶段：依据图片生成视频提示词

收到图片后检视实际外貌、服装、场景和道具，与已确认的设定对应。对应关系清楚就直接使用；图片缺失或有影响剧情的冲突时，只说明具体缺项或询问冲突项。不能把文字提示词当成实际参考图，也不提前输出待绑定的视频草案。

Read the full skill on GitHub

Details

Skill name: korean-thriller-asset-to-video
Repository: Work-Fisher/fisher-storylens
Repo updated: 15 October 2026

FAQ

What does the Korean Thriller Asset to Video skill do?

How do I install the Korean Thriller Asset to Video skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add Work-Fisher/fisher-storylens --skill korean-thriller-asset-to-video. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Korean Thriller Asset to Video popular?

It lives in Work-Fisher/fisher-storylens, which has 29 GitHub stars and 4 forks.

Related skills

Minimax H3 Creative Director

seraphinawolfe/minimax-h3-creative-agent-skill

Act as a creative director for MiniMax H3 video. Turns a goal such as an ad, a recipe short, a game teaser, a music clip or an explainer into a shot list with a consistent cast, writes H3-ready prompts, prices the whole piece, renders drafts and finals, and iterates on feedback. Use when the user wants a finished multi-shot piece rather than a single clip, or asks how to get consistent characters or products across MiniMax H3 shots.

Images, Video & Audio

Douban Sync

cosformula/douban-sync-skill

Export and sync Douban (豆瓣) book/movie/music/game collections to local CSV files (Obsidian-compatible). Use when the user wants to export their Douban reading/watching/listening/gaming history, set up incremental sync via RSS, or manage their Douban data locally.

Images, Video & Audio

Video Storyboard Gen

anyhui/video-storyboard-gen-skill

为中文视频创作生成可执行的故事结构、逐镜分镜表和即梦/Seedance 等 AI 视频提示词。用户提到分镜脚本、故事板、镜头设计、短视频/广告/宣传片脚本、运镜拆解、图生视频或文生视频提示词时使用；也适合把主题、文案或创意改造成带时间码的拍摄方案。不用于实际调用视频平台、规避平台规则，或编造产品功效与素材授权。

Images, Video & Audio

Read Anything

heyuxuan0209/read-anything

即时分析：任意链接（网页/博客/YouTube/B站/小宇宙播客/X）或本地录音（会议录音/语音备忘录）→ 抓取或本地转写 → 中文结构化解读（精读稿/快扫/访谈拆解/会议纪要/论文拆解/选题素材六种模板+自定义）→ 基于材料的多轮问答。触发：用户丢链接/音频文件/粘贴纪要并说"分析/解读/精读/这是什么/讲了啥/值得看吗/提选题"，或明确说"即时分析"。

Images, Video & Audio

Founder Voice Ghostwriter

BayramAnnakov/founder-voice-ghostwriter

Collaborative ghostwriting for founders and experts. Extracts first-hand knowledge through structured interviews and transforms it into authentic, non-AI-sounding content. Use when user wants to write thought leadership, blog posts, founder stories, comparison pages, or says things like "help me write", "make this authentic", "rewrite in my voice", or mentions avoiding AI detection.

Images, Video & Audio10

Zackd Director

Anil-matcha/zack-d-films-ai-video-generator

Turn ONE curiosity-driven topic or question into a finished Zack D Films-style 3D animated short video, end to end on the MuAPI platform + local ffmpeg — curiosity-loop script, character consistency sheets, 3D keyframe rendering, motion clips, cloned voiceover, zoom-ins, transitions, and captions, all automated. Use this whenever the user wants a "Zack D Films style short", a 3D medical / anatomical / educational explainer short, a curiosity-driven short video, or wants to automate 3D animated short production with an API-backed media pipeline. Triggers: "zack d films short", "zack d style", "

Images, Video & Audio

Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.