Video
Pack № 08 · Video
AI Video
Direct real video models from your terminal, with every paid step labelled.
npx skills add smixs/visual-skills@61a60d317f0fa0c3dc98902e58a7cb178838ae3f --skill video -g -y && \ npx skills add google-gemini/gemini-skills@c4d0c62c1240e08a90546186bf320600ab6a039a --skill gemini-omni-flash-api -g -y && \ npx skills add elevenlabs/skills@97478d1b23b9a6795f85bf8b6e33a4677fa1f869 --skill music sound-effects -g -y && \ npx skills add kajisho5/ffmpeg-skill@41a1e0c0c6185eccaf6a3bb1992a7ea9dece5b17 --skill ffmpeg-skill -g -y
npx skills add smixs/visual-skills@61a60d317f0fa0c3dc98902e58a7cb178838ae3f --skill video -g -y npx skills add google-gemini/gemini-skills@c4d0c62c1240e08a90546186bf320600ab6a039a --skill gemini-omni-flash-api -g -y npx skills add elevenlabs/skills@97478d1b23b9a6795f85bf8b6e33a4677fa1f869 --skill music sound-effects -g -y npx skills add kajisho5/ffmpeg-skill@41a1e0c0c6185eccaf6a3bb1992a7ea9dece5b17 --skill ffmpeg-skill -g -y
Fetches 5 skills from 4 author repos, each pinned to the version we reviewed. Nothing is mirrored. Uses the open-source skills CLI; needs Node.js 22.20 or newer.
Opens Claude Code with the install prompt already typed. Nothing runs until you press Enter.
Skills installed from a terminal do not reach claude.ai. A one-step route for claude.ai is in the works.
Until then you can add skills one at a time: open a card below, follow the link to the author's repo, and upload that skill folder as a ZIP under Customize, then Skills.
What's inside
5 skills. Each card links to its author's repository at the exact version we reviewed.
- AI video director 1/5 Video prompts Ready Turns a scene or script into prompts, storyboards and shot lists for AI video models, with a dramaturgy check on every shot and model-specific syntax for Seedance, Kling and Veo. @smixs CC BY 4.0 Source smixs/visual-skills at 61a60d3
AI video director
Video models reward direction, not adjectives. This plans the scene as a director would and writes the prompt last, so fewer paid generations go to a vague idea.
- Gemini Omni Flash video 2/5 Official Video generation Needs account Generates, extends and edits video with Google's Gemini Omni 1.2 Flash through bundled Python scripts: text, image or reference clips in, video files up to 4K out. @google-gemini Apache-2.0 Source google-gemini/gemini-skills at c4d0c62
Gemini Omni Flash video
The generator. Google's own skill for its video model: one API key, scripts you can read, files saved to your disk, and no vendor CLI or sign-in step in between.
- ElevenLabs music 3/5 Official Music generation Needs account Generates instrumental tracks and songs with the ElevenLabs Music API, from a prompt or a section-by-section composition plan, and can score music to a video you upload. @elevenlabs MIT Source elevenlabs/skills at 97478d1
ElevenLabs music
A score you can steer. ElevenLabs' own skill writes a track from a prompt or from a section plan you edit, and can fit music to a finished clip.
- ElevenLabs sound effects 4/5 Official Sound effects Needs account Generates sound effects, ambience and interface sounds from a text description with the ElevenLabs API, with control over length, looping and how closely it follows the prompt. @elevenlabs MIT Source elevenlabs/skills at 97478d1
ElevenLabs sound effects
Footsteps, impacts and room tone are what make generated footage believable. Short generations with control over length and looping.
- ffmpeg-skill 5/5 Video editing Needs app Edits video and audio with local FFmpeg through 42 bundled scripts: cut, join, reframe, captions, silence removal, loudness and platform exports, each with a dry run and a check. @kajisho5 MIT Source kajisho5/ffmpeg-skill at 41a1e0c
ffmpeg-skill
Clips, music and effects still have to become one file. Free and local: join the shots, mix the sound, set loudness and export for each platform.
Good to know
What you bring
- Gemini API key (GEMINI_API_KEY); Google bills each generated video
- Python 3.10 or newer with the google-genai package
- ffmpeg and ffprobe
- ElevenLabs account and API key (ELEVENLABS_API_KEY); generation uses paid credits
- FFmpeg 6.1 or newer
- Python 3.10 or newer
Pick one, not both
- Generator: this pack uses Google's Gemini Omni skill. For Seedance, Kling or Veo through one account, fal-video-generation (fal-ai-community/skills) does the same job; its guide prices a job before it runs and waits for your yes. It needs fal's genmedia CLI or MCP server. Install one generator, not both.
- Sound: Gemini Omni writes an audio track with the picture. The two ElevenLabs skills are for a separate score and effects that you mix yourself. Leave them out if the built-in audio is enough.
- Every skill file read in full
- Pinned to a reviewed version
- Author and licence on every card
- Requirements and costs disclosed
- No sponsored skills inside
Nov 2026Reviewed