Repolode
Submit
Agents365-ai

Agents365-ai/video-podcast-maker

A coding-agent skill that turns a topic into a narrated 4K video podcast, covering research, script, TTS, Remotion rendering and publishing for Chinese and Western platforms.

View on GitHub
Stars
1,715
Forks
179
Open issues
2
Last commit
last month

What is it?

Video Podcast Maker is an automated pipeline, packaged as an agent skill, that produces narrated video podcasts from a topic. It runs research, narration script generation, local text-to-speech (Edge free, or Azure), Remotion composition, 4K rendering, background music mixing with FFmpeg, and per-platform publish info. The workflow is driven interactively by a coding agent such as Claude Code, Codex or OpenCode, and the user makes the creative decisions while the agent handles the technical steps.

The pipeline centers on a single narration script file that drives audio, subtitles, section timing and animation sync. Assets such as user files, stock media, AI-generated images or B-roll are tracked in a per-video manifest that records source and license. Paid AI generation requires explicit confirmation before it runs. Output is 3840×2160 with subtitles rendered in React through Remotion, and the project can also cut 9:16 vertical shorts from long-form sections.

The tool targets creators publishing to Bilibili, YouTube, Xiaohongshu, Douyin and WeChat Channels in zh-CN or en-US. It includes pronunciation dictionaries for Chinese polyphones and can learn visual style from reference videos or images.

Key features

  • 01Generates research notes, narration script, TTS audio and subtitles from a single topic
  • 02Local TTS via Edge (no key required) or Azure Speech
  • 03Renders 3840×2160 video with Remotion and mixes background music with FFmpeg
  • 04Tracks asset source and license in a per-video manifest; paid AI generation requires confirmation
  • 05Produces 9:16 vertical shorts from long-form sections
  • 06Global and per-project pronunciation dictionaries for Chinese polyphones

Use cases

  • Turning a technical or educational topic into a narrated explainer video
  • Producing bilingual zh-CN and en-US videos for Bilibili, YouTube or Douyin
  • Repurposing long-form video sections into vertical short clips
  • Automating the script-to-render workflow from inside a coding agent

Quick start

  1. 1

    Install the skill with the skills CLI

    npx skills add Agents365-ai/video-podcast-maker/skills/video-podcast-maker -g
  2. 2

    Install Python dependencies (Python 3.8+, Node.js 18+ and FFmpeg must also be installed)

    pip install -r skills/video-podcast-maker/requirements.txt
  3. 3

    Create a Remotion project, or reuse an existing one, then install its packages

    npx create-video@latest my-video-project
  4. 4

    Set TTS_BACKEND and any required API keys in your shell profile, then ask your coding agent to create a video podcast

Things to know

  • The project is still under active development and may not be fully mature.
  • A fresh Remotion project downloads about 2.2 GB of npm packages plus a Chrome headless shell.
  • Azure TTS requires an Azure Speech key; the free Edge backend is the default.

In collections

Claude Code tools, skills & pluginsAI agent skill librariesAI video & audio tools

Topics

#ai-video#text-to-speech#remotion#video-generation#agent-skill#claude-code#podcast#bilibili#idea-to-video

FAQ

What is video-podcast-maker?
A coding-agent skill that turns a topic into a narrated 4K video podcast, covering research, script, TTS, Remotion rendering and publishing for Chinese and Western platforms. Video Podcast Maker is an automated pipeline, packaged as an agent skill, that produces narrated video podcasts from a topic. It runs research, narration script generation, local text-to-speech (Edge free, or Azure), Remotion composition, 4K rendering, background music mixing with FFmpeg, and per-platform publish info. The workflow is driven interactively by a coding agent such as Claude Code, Codex or OpenCode, and the user makes the creative decisions while the agent handles the technical steps.
Is video-podcast-maker free to use? What license does it have?
video-podcast-maker is open source under the MIT license. Check the license terms before using it in commercial products.
Is video-podcast-maker still maintained?
Updated within 6 months. The last commit to video-podcast-maker was on Oct 1, 2026.
What can I use video-podcast-maker for?
Turning a technical or educational topic into a narrated explainer video Producing bilingual zh-CN and en-US videos for Bilibili, YouTube or Douyin Repurposing long-form video sections into vertical short clips Automating the script-to-render workflow from inside a coding agent
How do I install video-podcast-maker?
Install the skill with the skills CLI: npx skills add Agents365-ai/video-podcast-maker/skills/video-podcast-maker -g Install Python dependencies (Python 3.8+, Node.js 18+ and FFmpeg must also be installed): pip install -r skills/video-podcast-maker/requirements.txt Create a Remotion project, or reuse an existing one, then install its packages: npx create-video@latest my-video-project Set TTS_BACKEND and any required API keys in your shell profile, then ask your coding agent to create a video podcast
Where can I get video-podcast-maker?
The source code and installation instructions are on GitHub: https://github.com/Agents365-ai/video-podcast-maker

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects