- Home
- Skills
- Workflow Automation
- Vertical AI Video Generator (Free Stack)
Works with the AI Agents you already use
Vertical AI Video Generator (Free Stack)
Turn a topic into a narrated vertical MP4 (1080x1920) with captions, music and SFX — 100% free stack, no GPU
Free
Vertical AI Video Generator (Free Stack)
Example session with this skill installed
Make me a short vertical video about '3 common mistakes when training calisthenics' — narrated in Spanish, anime-style visuals, no paid services.
- Read your context and instructions
- Compiled the vertical ai video
Summary
Built the full pipeline and generated output/calisthenics.mp4 — a 42-second, 1080x1920, 30fps MP4 with neural narration, word-level captions, background music and SFX. Total cost: $0.
What the agent produced
| Step | Tool | Result |
|---|---|---|
| Script | Groq (llama-3.1-8b) → validated JSON | 4 scenes, 2-8s each |
| Narration | edge-tts (es-MX-DaliaNeural) | narrator.mp3 + VTT → word captions |
| Images | Pollinations flux, seeded | 4 vertical backgrounds (cached per scene) |
| Render | Remotion + ffmpeg | H.264/AAC MP4, yuv420p |
// output/calisthenics.json — the manifest written next to the MP4
{
"title": "3 common mistakes when training calisthenics",
"style": "anime",
"provider": "groq",
"narrationEngine": "edge-tts",
"scenes": 4
}
Next steps
- Tweak
--voice,--rateor--style(anime / realistic / fantasy / all) and re-render - Run
--offlineon an air-gapped box — same flow, local placeholders + timed captions
Connects securely to your tools. The creator never sees your data.
About this skill
Turn any topic into a narrated vertical MP4 (1080x1920, 30fps) — script, neural narration, AI images, word-level captions, music and SFX — using only free services. No GPU, no credit card, and it degrades gracefully to fully offline output.
What this skill does
Guides your agent to build and run a complete short-form video pipeline:
- Script — LLM (Groq
llama-3.1-8b-instantor Geminigemini-2.0-flash-lite, both free tiers) generates a 3-6 scene script as strict JSON, validated with zod. A deterministic heuristic generator takes over automatically if the provider is down. - Narration —
edge-tts(free neural voices, e.g.es-MX-DaliaNeural) produces MP3 + VTT subtitles, converted to word-level captions. Offline fallback: timed silent audio via ffmpeg. - Images — Pollinations (
model=flux, 1080x1920, deterministic seed per prompt, no API key). Local per-style placeholders when it's unreachable. - Render — Remotion composition: camera motion per scene (static / slow pan / impact shake), music bed, per-scene SFX, highlighted captions, progress bar. Output: H.264/AAC MP4 + a JSON manifest making every render reproducible.
Includes
- The full pipeline design with the exact API calls, timeouts and validation rules that make it reliable
- Visual style presets (
anime,realistic,fantasy) — one command renders all three heuristic/--offlinemodes so the pipeline never hard-fails on a service outage- Low-memory render settings that work on free-tier boxes
Requirements
Node.js 20+, ffmpeg/ffprobe. Optional: Python 3 + edge-tts for neural narration; GROQ_API_KEY or GEMINI_API_KEY for LLM scripts (both have free tiers). Everything works without them.
Built and battle-tested in production on nova-studio.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
Requires Node 20+ and ffmpeg/ffprobe on PATH. No GPU acceleration; uses CPU-based headless Chrome rendering. Narration requires edge-tts (Python) for neural voices.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 26 days ago
- Free with an Agensi account
- Software
Node.js 20+FFmpegffprobePollinations image API (free, no key)
- Optional
Python 3edge-tts