跳到正文
TakoAPI
返回技能列表

ai-video-gen

由 rhanbourinajd 提供

Image & Video Generation
4.1k下载量由 clawskills.sh 提供
3星标
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

End-to-end AI video generation - create videos from text.

描述

该技能的功能

Generates complete videos from text prompts by chaining image generation, video synthesis, voice-over, and FFmpeg editing into a single pipeline. Supports multiple providers at each stage, with free and paid options. Scripts handle each step independently or as a full end-to-end run.

Combines image generation, video synthesis, and voice-over into one scriptable pipeline instead of manually moving files between separate tools.

何时使用

  • Generating a product demo video from a text description
  • Adding AI narration to an existing video clip
  • Turning a set of AI-generated images into an animation
  • Creating a multi-scene short video from scene-by-scene prompts
  • Producing social content with synced visuals and voice-over

示例工作流

输入

User asks: Generate a 5-second video of a futuristic city at night with a voice-over saying 'Welcome to the future'

智能体
  1. 1

    Runs generate_video.py with the prompt and voice-over text

  2. 2

    Generates frames using DALL-E 3 via OPENAI_API_KEY

  3. 3

    Synthesizes video from frames using LumaAI via LUMAAI_API_KEY

  4. 4

    Generates voice-over audio using OpenAI TTS

  5. 5

    Assembles video and audio into final MP4 using FFmpeg

输出

future_city.mp4 with synchronized visuals and narration

前置要求

  • OPENAI_API_KEY environment variable (for DALL-E 3 and/or OpenAI TTS)
  • REPLICATE_API_TOKEN environment variable (for Stable Diffusion or Flux via Replicate)
  • LUMAAI_API_KEY environment variable (for LumaAI Dream Machine)
  • RUNWAY_API_KEY environment variable (for Runway ML)
  • ELEVENLABS_API_KEY environment variable (for ElevenLabs voice-over)

安装

clawhub install rhanbourinajd/ai-video-gen

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接