<Devsly/>
NotebookLM Watermark Remover
Remove the NotebookLM watermark from PDF slide exports — free, unlimited, and 100% in your browser.
View
NotebookLM Video Watermark Remover
Remove the NotebookLM watermark from every frame of your MP4 and auto-trim the end card — free and unlimited.
View
Gemini Watermark Remover
Remove the Gemini (Veo) sparkle watermark from every frame of your AI-generated MP4 — free and unlimited.
View
MCP https://devsly.io/mcp MCP Tools Best Skills API Docs

Keyboard Shortcuts

Navigation

Quick Search
⌘ / Ctrl + K
Toggle Dark Mode
⌘ / Ctrl + D
Close Dialogs
Esc

Tool Actions

Execute / Process
⌘ / Ctrl + Enter
Clear Input
⌘ / Ctrl + Shift + C
Copy Output
⌘ / Ctrl + Shift + X

Editor

Select All
⌘ / Ctrl + A
Undo
⌘ / Ctrl + Z
Redo
⌘ / Ctrl + Shift + Z

Updated 1mo ago

by L-Trunks

ai-film-skills

本地 AI 短片生产线 skill 套件 —— 一张 12GB 显卡跑通 LTX / Wan / MiniMax H3,含 1600+ 行踩坑经验与防同质化机制

Works with
  • Claude

25 GitHub stars (25)

+4 in 16 days

Stars gained per day · daily snapshots since Sep 30
Stars gained per day, Sep 30 to Oct 16: +4 in total; the busiest day was Oct 2 with +1 stars.Oct 1: 0 starsOct 2: +1 starsOct 3: 0 starsOct 4: 0 starsOct 5: 0 starsOct 6: 0 starsOct 7: 0 starsOct 8: 0 starsOct 9: 0 starsOct 10: +1 starsOct 11: 0 starsOct 12: +1 starsOct 13: 0 starsOct 14: +1 starsOct 15: 0 starsOct 16: 0 stars
Stars gained between daily snapshots
PeriodNew stars
Oct 10
Oct 2+1
Oct 30
Oct 40
Oct 50
Oct 60
Oct 70
Oct 80
Oct 90
Oct 10+1
Oct 110
Oct 12+1
Oct 130
Oct 14+1
Oct 150
Oct 160
  • 6 skills (6 skills found in its SKILL.md files)
  • 5 forks (5 forks)
  • 1mo ago last update (Last updated 1mo ago)

Install

Skills CLI

npx skills add L-Trunks/ai-film-skills

Asks which skills and agents to install. Just one: npx skills add L-Trunks/ai-film-skills --skill emotion-to-camera-language — add -g for every project.

Claude Code

git clone https://github.com/L-Trunks/ai-film-skillscp -r "ai-film-skills/skills/emotion-to-camera-language" ~/.claude/skills/

Copies the emotion-to-camera-language skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.

Plugin

/plugin marketplace add L-Trunks/ai-film-skills

Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.

Skills can run code on your machine — read the SKILL.md first.

What it does

In the author's own words, from each skill's SKILL.md.

  • emotion-to-camera-language

    当用户要为 AI 生图/生视频写画面提示词,却只说得出「氛围感」「电影感」「高级感」「唯美」「治愈」「压抑」这类情绪形容词时调用。 也在用户说「出来的图很平/没感觉/像证件照/不像电影」而说不清哪里不对时调用。 本 skill 把情绪翻译成四个可观测变量:光位、景深、机位、人物状态。 不适用于:已经明确给出机位光位的具体提示词(无需翻译)、纯文字脚本创作、非视觉类需求、角色形象还没锁定(先用 lock-character-reference)。 Trigger: 氛围感 / 电影感 / 高级感 / 情绪片 / 写提示词 / 图很平 / 没感觉 / atmospheric / cinematic / mood

    skills/emotion-to-camera-language/SKILL.md on GitHub (opens in a new tab)
  • local-ai-film

    用本地或云端模型生产 AI 短片(氛围片/情绪片/预告片/vlog/带对白的剧情片)。导演总控入口:管开片决策、反同质化、踩坑规避、执行流程、后期总装。 触发场景:「做一部短片」「生成一个 XX 风格的视频」「跑一批氛围视频」「用 LTX/Wan/H3/可灵 出片」「AI 短片」「AI 剧情片」。 风格、时长、结构每次由需求决定,不预设。模型参数由 profile 决定,不写死。 不适用于:单张配图、已有实拍素材的剪辑。 Produce AI short films with local or cloud models. Directing hub: shot decisions, anti-homogenization, pitfall avoidance, execution, post. Trigger: "make a short film" / "generate an atm…

    skills/local-ai-film/SKILL.md on GitHub (opens in a new tab)
  • lock-character-reference

    **基准角色形象图还没跑出来、或反复跑不稳定**时调用——连一张能当参考的固定形象图都还没定下来。角色数 ≥3、或要换机位/景别/动作保持同一个人时也调用。 **分界**:基准图尚未确立 → 本 skill;基准图已稳、只是批量出图里某几张人物跑偏 → shot-breakdown。 本地 Qwen-Image-Edit 只能在同一张底图改衣服/风格,换机位换景别即失效,角色数应压到 1-2 个。 不适用:形象已锁定只问单镜提示词(emotion-to-camera-language)、云端平台有原生一致性时。 Trigger: 角色形象没定下来 / 参考图跑不稳 / 换个角度就变脸 / lock character

    skills/lock-character-reference/SKILL.md on GitHub (opens in a new tab)
  • material-driven-storyboard

    当用户在本地算力/工具受限的条件下写分镜,已经想好一个理想镜头(如「雨中奔跑」「追车戏」),却发现没有对应驱动视频或本地工作流做不出来时调用。 核心判断:分镜起点应是「手上真实有什么素材/驱动视频」,再倒推能写什么镜头,而不是先定理想画面再找办法实现。 提供三条出路:换驱动视频改动作 / 改用静态图加 Ken Burns / 改故事去掉该镜头。 不适用于:用户走云端平台(即梦/可灵)算力充足时——应转常规「理想画面先行」路线;单镜提示词怎么写用 emotion-to-camera-language。 Trigger: 素材反推 / 没有驱动视频 / 这个镜头做不出来 / driving video / material-driven

    skills/material-driven-storyboard/SKILL.md on GitHub (opens in a new tab)
  • rhythm-density

    当用户有多个本地生成的短镜头,要决定"每镜多长、怎么排布"才有节奏感时调用;尤其想做"长镜头"呼吸感却受限于单镜时长上限时。核心:高能段压到疏密比的短端快切,首尾放慢做呼吸;单镜时长上限由 profile 的 max_shot_sec 决定,不是固定数值;上限之下用同机位连续两镜+极缓运镜拼接制造"伪长镜"。 不适用于:音效/调色/BGM决策(见 editing-triad)、生成阶段镜头筛选(见 shot-breakdown)、单镜头光位机位描述(见 emotion-to-camera-language)。 也不适用于旁白驱动的科普/教程片剪辑刀数决策(见 shot-rhythm-model)——那个模型服务有解说词的信息片,节奏系数由概念/演示/验收等内容段类型决定;本 skill 服务无对白情绪片,没有旁白锚点,节奏服务情绪起伏且受单镜时长上限收敛,两套参数不能互相套用。 Trig…

    skills/rhythm-density/SKILL.md on GitHub (opens in a new tab)
  • shot-breakdown

    角色基准形象图**已锁定**后,批量出图并从结果里筛可用镜头时调用。典型:矩阵化生图后挑符合脚本意图的;某张景别/机位不对要重抽;同镜反复抽不出来、要决定继续抽还是弃用改故事。 **分界**:基准形象图本身跑不稳 → lock-character-reference;基准图已稳、只是这批结果里人物跑偏或景别不对 → 本 skill。 本地每次生成分钟级,同镜重抽 3-5 次不可用就停手改故事。 不适用:生成前提示词怎么写(emotion-to-camera-language)、镜间时长排布(rhythm-density)。 Trigger: 矩阵生图 / 筛图 / 重抽 / 景别不对 / shot breakdown

    skills/shot-breakdown/SKILL.md on GitHub (opens in a new tab)

Review before you install

A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.