agentleFS
Dimks777Dimks777/aiclub/skills/timelapse-creator/SKILL.md

timelapse-creator

Skill0 starsChanged 7 months agoMITView on GitHub

Создание AI-таймлапсов от идеи до готовых клипов. Активируется на: таймлапс, timelapse, переход между кадрами, трансформация, видео-переход.

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When user says "давай таймлапс" or similar — brainstorm the concept:

    line 27
  2. Distinctive:

    Convert PNG→JPEG: before sending (4K PNG > 8MB → 413 error)

    line 44
  3. Distinctive:

    If approved → Phase 4. If not → regenerate specific frames.

    line 69
  4. Distinctive:

    Use base64 encoding (no need for public URL hosting)

    line 79

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Timelapse Creator

Generate timelapse-style videos through a structured pipeline: idea → keyframes → approval → video clips → delivery.

Pipeline Overview

IDEA → KEYFRAMES (Nano Banana) → APPROVAL → VIDEO (Veo 3.1-fl) → DELIVER mp4

Phase 1: Ideation

When user says "давай таймлапс" or similar — brainstorm the concept:

  • What transforms? (object, room, vehicle, food, etc.)
  • From what state to what state? (broken→fixed, empty→decorated, raw→cooked)
  • How many intermediate steps? (usually 4-5 total keyframes)
  • Setting/environment? (workshop, kitchen, studio, outdoor)

Propose 4-5 keyframe descriptions, get approval before generating.


Phase 2: Keyframes (Nano Banana Pro)

Rules

  1. ANCHOR PROMPT — write once, reuse for ALL frames. Describes: camera angle, location, lighting. Must be identical.
  2. VARIABLE PART — unique per frame. Describes the current state of transformation.
  3. Format: 9:16 vertical, 2K (or 4K if requested)
  4. Prompts ALWAYS in English
  5. Convert PNG→JPEG before sending (4K PNG > 8MB → 413 error)
  6. Save all frames to content/timelapse/{project-name}/

Template

[ANCHOR — identical for all frames]
Interior of a workshop, [object] centered in frame, shot from 
[angle] at [distance]. [Lighting description]. Same setup throughout.

[VARIABLE — unique per frame]  
Frame 1: The [object] shows [damaged state description]...
Frame 2: The [object] has been [intermediate state]...
Frame 3: The [object] is being [process description]...
Frame 4: The [object] is [final state description]...

Phase 3: Approval

Send all keyframes to user. Check:

  • Consistency (same scene, same angle?)
  • Logical progression (does transformation make sense?)
  • Quality (any artifacts, weird details?)

If approved → Phase 4. If not → regenerate specific frames.


Phase 4: Video (Veo 3.1-fl)

Используй скилл veo-video для генерации видео.

Rules

  1. First image = START frame, second = END frame
  2. Use base64 encoding (no need for public URL hosting)
  3. Generate pairs: frame1→frame2, frame2→frame3, frame3→frame4
  4. Prompt describes the PROCESS of transformation
  5. Model: veo-3.1-fast-fl for tests, veo-3.1-fl for final

Video Prompt Template

Smooth realistic timelapse of [process description]. Fixed camera angle, 
no camera movement. Continuous physical transformation, realistic lighting 
changes. No dissolves, no crossfades — only physical build-up and assembly.

Request Format (Sync API)

{
  "messages": [{"role": "user", "content": [
    {"type": "text", "text": "Smooth timelapse transition..."},
    {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,{START}"}},
    {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,{END}"}}
  ]}],
  "model": "veo-3.1-fast-fl",
  "stream": true
}

Phase 5: Delivery

  1. Save all mp4 clips to content/videos/{project-name}/
  2. Send to client
  3. Client assembles in CapCut (speeds up 8s→2-3s per segment)

Cost Summary

ItemQtyPriceTotal
Keyframes4-5~$0.05$0.20-0.25
Video clips3-4~$0.15$0.45-0.60
TOTAL~$0.70-0.85

Transition Prompt Templates

Time of Day

Smooth transition from day to night, realistic lighting changes,
maintaining the same camera angle and composition

Construction/Growth

Timelapse showing gradual construction progress,
realistic materials appearing, smooth continuous motion

Nature Changes

Natural progression showing seasonal change,
realistic weather transition, maintaining landscape composition

Space Transformation

Smooth transformation of the space,
realistic material changes, maintaining perspective and scale

Troubleshooting

ПроблемаДействие
Перепутал API ключиNano = кадры, Veo = видео. Разные ключи!
Видео не соответствует кадрамПроверить суффикс -fl в модели
Переход неестественныйПеределать кадры — больше сходства
Качество низкоеПереключиться с fast-fl на fl
API ошибкаПроверить размер кадров < 8MB

Video Output Specs

ПараметрЗначение
Длительность5-10 секунд
КачествоHD/4K
ФорматMP4
FPS24-30
Кадры входJPG < 8MB, идентичный aspect ratio

Output Structure

content/timelapse/{project-name}/
├── frames/
│   ├── frame-1.jpg
│   ├── frame-2.jpg
│   ├── frame-3.jpg
│   └── frame-4.jpg
├── videos/
│   ├── segment-1-2.mp4
│   ├── segment-2-3.mp4
│   └── segment-3-4.mp4
├── prompts.md
└── notes.md

Timelapse Creator v1.0 — Content Factory

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.