muapi-cinema-director
Direct high-fidelity cinematic video with AI — translates creative intent into technical cinematographic directives for Veo3, Kling, and Luma video models via muapi.ai
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-11
- Stars
- ★ 1.6k
- Forks
- 474
01 前言 Front Matter
- name
- muapi-cinema-director
- version
- 0.1.0
- description
- Direct high-fidelity cinematic video with AI — translates creative intent into technical cinematographic directives for Veo3, Kling, and Luma video models via muapi.ai
02 资源文件 1
- 脚本scripts/generate-film.sh2.4 KB
03 SKILL.md 正文
? AI Cinema Director Skill
A specialized skill for AI Agents to direct high-fidelity cinematic video. The Cinema Director skill translates high-level creative intent into technical cinematographic directives for state-of-the-art video models (Veo3, Kling, Luma).
Core Competencies
- Shot Composition Analysis: Mapping emotional beats to appropriate framing (e.g., Extreme Close-Up for intimacy, Wide Shot for isolation).
- Camera Movement Orchestration: Directing complex physical movements (Dolly, Truck, Crane) and lens-based effects (Rack Focus, Dolly Zoom).
- Lighting & Atmosphere Design: Specifying temporal and stylistic lighting (Golden Hour, Chiaroscuro, Volumetric God Rays).
- Technical Parameter Optimization: Automatically selecting optimal frame rates, aspect ratios, and model-specific biases.
?️ Technical Specification
1. Intent Mapping Table
| Creative Intent | Framing | Movement | Lighting |
|---|---|---|---|
| Heroic Reveal | Low Angle / Wide | Crane Up / Orbit | Rim Lighting / High Contrast |
| Tense/Uneasy | Dutch Angle | Handheld Shake | Low Key / Harsh Shadows |
| Introspective | Close-Up | Slow Push In | Soft Rembrandt / Window Light |
| Majestic/Epic | Extreme Wide | Drone Flyover | Golden Hour / Volumetric |
| Melancholic | Profile / Medium | Slow Pull Out | Blue Hour / Desaturated |
2. Physical Camera Movements
Dolly In/Out: Physical camera movement on a track toward/away from the subject.Truck Left/Right: Lateral physical movement.Crane/Jib: Sweeping vertical movement from a height.Orbit: Circular movement around a center point.Pedestal: Vertical elevation change (without tilting).
3. Lens & Optical Controls
Shallow DOF: Background blur (Bokeh).Anamorphic: Horizontal flares and wide-screen cinematic feel.Rack Focus: Shifting focus between planes within the shot.
? Prompt Optimization Protocol (Agent Instruction)
Before calling the script, the Agent MUST expand the user's raw prompt using these Director's Rules:
- TECHNICAL INFUSION: Transform "cool action" into a technical shot.
- Formula:
[Shot Type] + [Subject/Action] + [Environment] + [Lighting] + [Camera Movement] + [Lens Effect]
- Formula:
- MOTION DYNAMICS: Use cinematic verbs: Dolly In (intimacy), Crane Up (majestic), Orbit (heroic), Truck (parallel motion).
- LIGHTING RECIPES: Apply specific illumination: Volumetric God Rays, Teal-and-Orange Grade, Cyberpunk Rim Lighting, Rembrandt Portrait Lighting.
- PHYSICS LOGIC: Describe light relationships (e.g., "Neon reflections shimmering on rain-slicked asphalt") to trigger model reasoning.
? Protocol: Using the Cinema Director
Step 1: Define the Creative Brief
Provide the agent with a subject and a "Director's Intent."
Step 2: Invoke the Script
The generate-film.sh script accepts a --brief which it expands using its internal knowledge of cinematography.
# Directing a scene
bash scripts/generate-film.sh
--subject "A lone samurai in a blizzard"
--intent "epic reveal"
--model "kling-master"
Step 3: Handle the Async Response
Video generation is asynchronous. Use the returned request_id to poll for completion via core/platform/check-result.sh.
⚠️ Constraints & Guardrails
- Temporal Consistency: Avoid complex subject transformations in a single shot (e.g., "man turns into a bird").
- Movement Collisions: Do not combine contradictory movements (e.g., "Dolly In" and "Dolly Out" simultaneously).
- Physical Realism: Prefer movements possible with real-world equipment for a more professional "film" look.
- Model Bias:
Veo3: Best for slow, high-quality aesthetic shots.Kling: Best for complex character motion and physics.Luma: Best for fast-paced, high-action cinematic sequences.
⚙️ Implementation Details
This skill acts as an "Expert Translator" for the core/media/generate-video.sh primitive. It maintains a dictionary of cinematic styles and injects technical directives into the prompt before execution.
04 同仓库更多 Skills
migrate-to-e2e
Moves an existing Playwright, Cypress, Selenium, Appium, Detox, Maestro, or TesterArmy suite to e2e one slice at a time, with the old suite running beside the new one until each ported test is proven. Keeps a MIGRATION.md progress file so any later session resumes where the last stopped. Optionally plants a defect to show both suites catch it. Use for /migrate-to-e2e, "migrate our tests to e2e", "port these Maestro flows", "move our TesterArmy tests", or when a project has an e2e config and an older suite still running.
e2e
Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.
create-verification-skill
Generate a project-local verify-<app> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for /create-verification-skill, "make a verification skill for this repo", or when a project has no scripted way to prove UI behavior.
deja-search
deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.
deja-reflect
End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.
revid
Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.