cclank/lanshu-create-ai-presenter-video
Code reviewThis page was translated automatically. Technical details may differ from the original. View the original
Provider-neutral Codex Skill for producing verified AI presenter videos from a script and an authorized presenter image.
Overview
A universal digital human video production skill for Codex. After entering a theme or copy and authorized character images, it organizes the copywriting, voiceover, character generation, lip-sync calibration, subtitles and keyword animations, editing, rendering, and quality acceptance. Optional inputs include voice samples, screen recording, images, B-roll, brand assets, target platform, duration, landscape and portrait, style, watermark, and end guidance. - generation.md: Input check, copywriting, voice, ability selection, paid generation, character prompts and consistency. - editing.md: Timeline, opening and ending, subtitle presets, character-side keyword animations, cover and export. - qa-recovery.md: Technical acceptance, manual acceptance, and common fault fixes. - Codex or an agent environment compatible with local skills. - Python 3.9+。 - FFmpeg and ffprobe. - Bash, jq, awk, and sed. - At least one video generation, speech generation, and lip-sync capability that can be called in the current environment. Full dubbing is the timeline for the entire film. Character videos, subtitles, shots, keywords, and transitions are all positioned according to this audio line, which helps reduce lip-sync drift and segment transitions. - Portrait 9:16, 1080×1920, 30fps. - Theme-generated videos are usually controlled between 45–75 seconds. - Use appropriate stock sounds when no authorized voice samples are available. - Default includes a clear opening, 2–4 content beats, and a concise ending. - Music and commercial lead can be added as needed. - Release loudness target is about -16 LUFS. - Confirm image usage rights and adult product status before remote upload. - Confirm voice authorization before cloning voices. - Explain uploaded content, generation duration, price basis, sample plan, and retry limit before generating the first payment. - Prioritize querying existing task IDs after task interruption to avoid duplicate deductions. - Stop and summarize issues after three consecutive payment candidates fail. - The repository does not store API keys, access tokens, signed download addresses, or user materials. - Task-level request records require removing credentials and temporary URLs before submitting. - Pretest and delivery reports only save file names, not the absolute path of the developer machine. - The project uses the MIT License, which can be freely used, modified, and distributed.
Overview (machine translated)
English | Chinese Simplified
Turn a topic or script into a verified, publish-ready explainer video — presented by a digital human from one authorized portrait, or acted out on screen in one of twelve explainer styles — driven by the coding agent you already use.
lanshu-create-ai-presenter-video is an Agent Skill that takes an AI agent through the full production of an explainer video. On the presenter route that means script, narration, presenter generation, lip-sync, captions and keyword motion graphics, editing, rendering, and quality assurance. On the styled route the narration is performed scene by scene in one of twelve styles (nine visual, three narrative), from a word-timed story to a rendered film. Every stage is gated by evidence on disk, and nothing is delivered until the output passes decode and loudness checks.
Highlights
- Harness-neutral. A standard
SKILL.mdthat loads in Claude Code, Codex, Gemini CLI, Cursor, OpenCode, GitHub Copilot, and other Agent Skills clients. Any agent that can read files and run shell commands can follow it directly. - **Provider-neutral.
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add cclank/lanshu-create-ai-presenter-video