video-production-skill
video-production-skill takes an agent from script to video: slides, speech-verified narration, subtitles and a cover.
- 106 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Texts and keys go to ElevenLabs and OpenAI, usage is billed at their rates.
- Scripts run FFmpeg and Playwright on your computer and write files into the working folder.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/video-production-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Texts and keys go to ElevenLabs and OpenAI, usage is billed at their rates.
- Scripts run FFmpeg and Playwright on your computer and write files into the working folder.
This is third-party code. Review the repository files before installing.
What it does
The skill describes a pipeline for narrated slide-based teaching videos. A script becomes slides (hand-drawn through `gpt-image-2`, or HTML captured with Playwright), ElevenLabs voices the text and Whisper checks every line, which must score at least 0.85. FFmpeg then assembles the video, subtitles are aligned word by word and a cover is prepared. The repository holds a `SKILL.md` with a mandatory checklist, nine scripts and references: teaching style, narration writing, a homograph table and lessons from more than forty videos. The author tuned the pipeline for Chinese (traditional script); for other languages you adjust recognition thresholds and subtitle width yourself.
Who it is for. Authors of educational channels and courses who want an agent to produce narrated slide videos.
Good fit when
- An agent makes a series of explainer videos using one method.
- Each narrated phrase has to be checked against the source text.
- Subtitles must match speech and not split English words.
- A subagent receives a make-a-video task and has to follow the checklist.
Not a fit when
- You need a generator for live-action footage and editing.
- You have no ElevenLabs and OpenAI keys, or texts cannot be sent to those services.
- A Russian-language video is needed without tuning thresholds and fonts.
Example request
Make a five-minute teaching video about Git basics with `video-production-skill`: script, slides, narration, subtitles and cover.Limitations
Requires Node.js 18 or newer, Python 3.9 or newer, FFmpeg with FFprobe, and the keys `ELEVENLABS_API_KEY` and `OPENAI_API_KEY`. A ten-slide video costs roughly ten to fifteen image generations, ten to twenty speech syntheses and twenty to thirty transcriptions. The project is tuned for Chinese. The last commit dates from July 2026. The author asks to disclose that a video was made with AI. Before running, check which texts go to external services and set spending limits in the providers' dashboards.
How to disable. Delete the skill folder from `.claude/skills/` (or the `SKILL.md` line from `AGENTS.md` in Codex). Revoke the ElevenLabs and OpenAI keys in their dashboards if no longer needed.
FAQ
Is this a text-to-video generator?
No. It is an instruction for an agent plus a set of scripts that assemble a video from slides, narration and subtitles.
How is narration verified?
Whisper transcribes every narrated line, and similarity to the source text must be at least 0.85.
Are paid services required?
Yes. Narration uses ElevenLabs, verification and images use OpenAI. Slides can be built in HTML without image generation.
What about another language?
The pipeline logic carries over, but recognition thresholds and subtitle width need tuning. The homograph table is written for Chinese.
What is produced in the end?
A video file with slides, narration and aligned subtitles plus a cover. Intermediate files remain in the working folder.
Discussion
No questions or reviews yet.
Related
SkillsA skill that turns a topic into a narrated 4K video: research, script, local TTS, Remotion composition and an FFmpeg mix
Skills for faceless video channels: one topic becomes a video with voiceover, subtitles, covers and per-platform copy.
A skill that turns an idea, a link or a screen recording into finished video: HeyGen avatar, b-roll, captions and quality checks
The video-explainer skill turns a script into a vertical 9:16 explainer: narration, captions, Remotion and a pre-render review
A Claude Code skill that records a product in the browser, composes a video with chapters and captions and masters the MP4 locally
Thirteen HyperFrames video skills: script, motion samples, A/B shot review, voiceover, captions and final checks
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| video-production-skill (this card) | no | 106 | |
| Video Production Skills | no | 710 | provides a different set of video skills; this one is a single pipeline with narration verified by recognition |
| Slides Maker | no | 667 | makes slides; video-production-skill adds narration, subtitles and video assembly |
README in short
The author published the working pipeline unchanged, swapping only personal voice IDs for config fields. The most valuable part is the lessons file from forty videos.