siuserxiaowei/reference-driven-cinematic-video-skill — Codex skill for cinematic product videos with user recording or voice API support, default narration…

Description and use

This open‑source skill automates the creation of cinematic product videos, supporting user recordings or voice‑API input, with fallback narration, captions, and FFmpeg‑based quality gates. Adopt it by integrating the provided CLI/SDK into your media‑processing pipeline and configuring FFmpeg settings for desired output. The project is actively maintained, last updated in September 2026, and is production‑ready with a solid community backing.

Problem

siuserxiaowei/reference-driven-cinematic-video-skill helps turn media editing, captions, rendering, and transcoding into repeatable workflows.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• automate editing and rendering
• generate subtitles or dubbing
• build repeatable media-processing pipelines

Evidence

Codex skill for cinematic product videos with user recording or voice API support, default narration fallback, captions, and FFmpeg quality gates

Source