← Back to Skills Marketplace
zhangifonly

Luma Ai

by zhangifonly · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
177
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install luma-ai
Description
Luma AI 视频生成助手,精通 Dream Machine 文生视频、图生视频、提示词技巧
README (SKILL.md)

Luma AI 助手

你是一个精通 Luma AI(Dream Machine)的 AI 助手,能够帮助用户生成高质量 AI 视频。

身份与能力

  • 精通 Luma Dream Machine 的文生视频和图生视频功能
  • 熟悉视频生成提示词技巧和镜头语言
  • 掌握 Luma API 调用方式
  • 了解与 Runway、Kling、Sora 等竞品的差异

核心功能

文生视频(Text to Video)

输入文字描述 → 生成 5 秒视频片段

  • 支持自然语言描述场景、动作、风格
  • 可指定镜头运动(推拉摇移)
  • 支持多种风格(写实、动画、电影感)

图生视频(Image to Video)

上传参考图 → 生成基于该图的动态视频

  • 保持图片风格和主体一致性
  • 适合让静态设计稿"动起来"

关键帧控制

设置起始帧和结束帧图片,AI 生成中间过渡动画。

提示词技巧

结构

[主体] + [动作] + [场景] + [镜头] + [风格/氛围]

示例

  • 基础:"一只橘猫在窗台上打哈欠,阳光洒进来,温暖的午后氛围"
  • 电影感:"cinematic shot, a woman walking through a rainy Tokyo street at night, neon reflections on wet pavement, slow motion"
  • 镜头运动:"drone shot slowly rising above a misty mountain forest at sunrise, golden hour lighting"

镜头语言关键词

关键词 效果
tracking shot 跟踪镜头
dolly zoom 推拉变焦
aerial/drone shot 航拍
slow motion 慢动作
timelapse 延时摄影
close-up 特写
wide angle 广角
handheld camera 手持摄影感

API 调用

import requests

response = requests.post(
    "https://api.lumalabs.ai/dream-machine/v1/generations",
    headers={"Authorization": "Bearer luma-xxx"},
    json={
        "prompt": "a cat sitting on a windowsill, warm sunlight",
        "aspect_ratio": "16:9",
        "loop": False
    }
)
generation_id = response.json()["id"]

# 轮询获取结果
result = requests.get(
    f"https://api.lumalabs.ai/dream-machine/v1/generations/{generation_id}",
    headers={"Authorization": "Bearer luma-xxx"}
).json()
# result["assets"]["video"] 为视频 URL

与竞品对比

维度 Luma Runway Gen-3 Kling Sora
运动自然度 优秀 优秀 良好 顶级
生成速度 中等 中等
免费额度 有限
API 有限
中文提示词 一般 一般 优秀 良好

最佳实践

  • 英文提示词效果通常优于中文
  • 描述要具体,避免抽象概念
  • 镜头运动关键词放在提示词开头效果更好
  • 图生视频时,参考图质量直接影响输出质量
  • 生成多个版本挑选最佳,AI 视频有随机性

最后更新: 2026-03-22

Usage Guidance
This skill appears to be a straightforward Luma/Dream Machine usage guide and includes example API calls, but it does not declare the API key/token it expects. Before installing or using it: 1) Confirm the skill's origin or request a homepage/repository — no source is provided. 2) Expect the agent or the skill to ask you for a Luma API key at runtime; do not paste long-lived or high-privilege keys into unfamiliar skills. Prefer creating a scoped, revocable API key for testing. 3) Verify the API host (api.lumalabs.ai) is correct and official. 4) If you need stronger assurance, ask the skill author to declare a required env var (e.g., LUMA_API_KEY) in the metadata and to provide clear guidance on least-privilege key scopes. These steps will reduce the risk that a skill silently exfiltrates credentials or misuses a broad API key.
Capability Analysis
Type: OpenClaw Skill Name: luma-ai Version: 1.0.0 The skill bundle is a standard documentation-based assistant for Luma AI video generation. It contains instructional content for prompt engineering and a legitimate API usage example for 'api.lumalabs.ai' in SKILL.md. No malicious code, data exfiltration, or prompt injection attacks were identified.
Capability Assessment
Purpose & Capability
The name/description (Luma AI video generation) align with the SKILL.md which documents text-to-video, image-to-video, keyframe control, and shows example API calls to api.lumalabs.ai. That purpose is coherent with the included code examples.
Instruction Scope
The SKILL.md contains runnable API examples that include an Authorization: Bearer luma-xxx header and polling for generation results. However, the instructions do not specify how the agent gets/uses the API credential (no declared env var or guidance for secure handling). The instructions otherwise do not reference unrelated files, system paths, or other credentials.
Install Mechanism
This is an instruction-only skill with no install spec and no code files. That minimizes on-disk installation risk.
Credentials
The SKILL.md clearly expects an API bearer token for api.lumalabs.ai, but the skill metadata lists no required environment variables or primary credential. That mismatch is disproportionate: a network API integration normally declares a required API key or token so the platform/user can review and control it.
Persistence & Privilege
The skill is not always-enabled and is user-invocable; it does not request elevated or persistent platform privileges in the metadata.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install luma-ai
  3. After installation, invoke the skill by name or use /luma-ai
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Luma AI Skill 1.0.0 – Initial Release - Introduces a comprehensive Luma AI (Dream Machine) assistant for AI-driven video generation. - Supports text-to-video, image-to-video, and keyframe animation features. - Provides detailed prompt engineering tips, including structure, examples, and camera language keywords. - Includes Luma API usage sample for easy video generation integration. - Compares Luma’s capabilities with major competitors (Runway, Kling, Sora). - Shares best practices for optimal AI video creation.
Metadata
Slug luma-ai
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Luma Ai?

Luma AI 视频生成助手,精通 Dream Machine 文生视频、图生视频、提示词技巧. It is an AI Agent Skill for Claude Code / OpenClaw, with 177 downloads so far.

How do I install Luma Ai?

Run "/install luma-ai" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Luma Ai free?

Yes, Luma Ai is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Luma Ai support?

Luma Ai is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Luma Ai?

It is built and maintained by zhangifonly (@zhangifonly); the current version is v1.0.0.

💬 Comments