agentsclimarketplace

Ai video director

Skill Sheshiyer/skill-clusters/skills/ai-video-director

Hub-and-spoke agent-skill clusters, one per stack (Astro·GSAP·Remotion, Tauri, …). Installable via skills.sh.

Install
npx -y skills add Sheshiyer/skill-clusters --skill ai-video-director

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Create AI-generated video content using Kling 3.0 (multi-shot sequences with dialogue), Sora (scene generation with character consistency), or Grok (restricted content). Use when writing shot-by-shot video prompts, building character-consistent animated series, or routing between video generation tools.

SKILL.md

3.4 KB, 699 tokens by cl100k_base, as published. Nobody here has run it

AI Video Director

Write shot-by-shot video prompt sequences for AI video generation tools. Supports multi-shot sequences (up to 6 shots), dialogue with voice control, and character consistency across scenes.

When to Use

  • Writing multi-shot video sequences with scene transitions
  • Creating AI-generated animated series with consistent characters
  • Adding dialogue with emotion/tone control to video prompts
  • Routing between video tools (Kling 3.0, Sora, Grok) based on content needs

Tool Routing

ToolBest ForKey Capability
Kling 3.0Multi-shot sequences with dialogueUp to 6 native shots, audio control, subject anchoring
SoraScene generation, character consistencyBest for narrative scenes, handles [cut] transitions
GrokRestricted/romantic contentFewer content restrictions than Sora
MidjourneyCharacter source imagesStyle/Omni reference for initial character design

Workflow

  1. Define characters -- Lock identity traits early (appearance, clothing, key features)
  2. Write shot sequence -- Describe each shot with cinematic language (profile, macro, tracking, POV)
  3. Add audio -- Structure dialogue as [Character: Label, tone]: "text"
  4. Route to tool -- Select based on content requirements and shot count
  5. Generate insert shots -- Create reusable clips for character consistency

Core Principles (Kling 3.0)

  1. Think in Shots, Not Clips -- Use cinematic language (profile shots, macro close-ups, tracking, shot-reverse-shot)
  2. Anchor Subjects Early -- Define core subjects clearly at the beginning for consistency
  3. Describe Motion Explicitly -- Specify camera behavior (tracking, following, freezing, panning) AND subject movement
  4. Native Audio -- Use [Character A: Label, tone]: "Dialogue" format with linking words for rhythm
  5. Image-to-Video: Lock First -- Treat input image as anchor, focus on how scene evolves FROM the image

Audio Prompting Rules

PrincipleGuidelineExample
Structured NamingUnique, consistent labels[Character A: Black-suited Agent]
Visual AnchoringAction first, then dialogueThe agent slams the table. [Agent, angrily]: "Where?"
Audio DetailsTone and emotion labels[Agent, raspy, deep voice]: "Don't move."
Temporal ControlLinking words for rhythm[Agent]: "Why?" Immediately, [Assistant]: "Because."

Character Consistency Workflow (Framer X)

For animated series with consistent characters across multiple scenes:

  1. Step 1: Generate character source images (Midjourney Style/Omni reference)
  2. Step 2: Create establishing scenes (Sora) with [cut] transitions
  3. Step 3: Generate interaction scenes (Sora) in different environments
  4. Step 4: Create insert shots for reusable character clips
  5. Step 5: Route restricted content to Grok with custom source images

See references/prompt-templates.md for complete examples.

What ships with it: 8 files

7622.4 KB alongside SKILL.md

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.