name: video-podcast-maker description: Use when the user gives a topic and wants an automated video podcast created, or asks to learn visual design patterns from a reference video/image. Produces 4K video via research → script → TTS → Remotion → MP4 + BGM. argument-hint: "[topic]" effort: high

allowed-tools intentionally omitted — the workflow uses Bash, Read, Write, Edit,

Glob, Grep, WebFetch, WebSearch, Agent, AND optional MCP tools (Playwright for

design learning). Listing a closed set here would silently break the Playwright

fallback. Defer tool gating to the user's Claude Code permission settings.

--- Frontmatter fields above are primarily for Claude Code / OpenClaw.

Other agents such as Codex should ignore unknown fields and follow the workflow below. ---

author: Agents365-ai category: Content Creation version: 2.0.0 created: 2025-01-27 updated: 2026-04-03 bilibili: https://space.bilibili.com/441831884 github: https://github.com/Agents365-ai/video-podcast-maker

dependencies is informational only — Claude Code does not auto-load skills

from this list. The agent must invoke remotion-best-practices per the body

section below. Kept for OpenClaw / SkillsMP discovery metadata.

dependencies:

• remotion-best-practices metadata: openclaw: requires: bins: - python3 - ffmpeg - node - npx primaryEnv: AZURE_SPEECH_KEY emoji: "🎬" homepage: https://github.com/Agents365-ai/video-podcast-maker os: ["macos", "linux"] install:
  • kind: brew formula: ffmpeg bins: [ffmpeg]
  • kind: uv package: edge-tts bins: [edge-tts]

REQUIRED: Load Remotion Best Practices First

This skill depends on remotion-best-practices. You MUST invoke it before proceeding:

Invoke the skill/tool named: remotion-best-practices

Video Podcast Maker

Quick Start

Open your coding agent and say: "Make a video podcast about $ARGUMENTS"

Or invoke directly: /video-podcast-maker AI Agent tutorial

Design Learning (Optional)

Extract visual design patterns from reference videos or images and apply them to new video compositions. Skip this section unless the user provides a reference video/image or asks to save/list/delete style profiles.

→ See references/design-learning.md for commands, reference-library management, style-profile management, and integration with Pre-workflow / Step 9.

Auto Update Check

Agent behavior: Check for updates at most once per day (throttled by timestamp file). Before any shell command that reads files from this skill, resolve SKILL_DIR to the directory containing SKILL.md. If your agent exposes a built-in skill directory variable such as ${CLAUDE_SKILL_DIR}, you may map it to SKILL_DIR.

SKILL_DIR="${SKILL_DIR:-${CLAUDE_SKILL_DIR}}"
STAMP="${SKILL_DIR}/.last_update_check"
NOW=$(date +%s)
LAST=$(cat "$STAMP" 2>/dev/null || echo 0)
if [ ! -d "${SKILL_DIR}/.git" ]; then
  echo "MANUAL_INSTALL"
elif [ $((NOW - LAST)) -gt 86400 ]; then
  timeout 5 git -C "${SKILL_DIR}" fetch --quiet 2>/dev/null || true
  LOCAL=$(git -C "${SKILL_DIR}" rev-parse HEAD 2>/dev/null)
  REMOTE=$(git -C "${SKILL_DIR}" rev-parse origin/main 2>/dev/null)
  echo "$NOW" > "$STAMP"
  if [ -n "$LOCAL" ] && [ -n "$REMOTE" ] && [ "$LOCAL" != "$REMOTE" ]; then
    echo "UPDATE_AVAILABLE"
  else
    echo "UP_TO_DATE"
  fi
else
  echo "SKIPPED_RECENT_CHECK"
fi

• Update available: Ask the user whether to pull updates. Yes → git -C "${SKILL_DIR}" pull. No → continue.
• Up to date / Skipped: Continue silently.
• Manual install (no .git directory — skill was installed via tarball/zip/cp): Continue silently. Auto-update is disabled; the user must reinstall manually to update.

Prerequisites Check

Agent behavior: Run the pre-flight check below before Step 1. SKILL_DIR must already be resolved (see Auto Update Check section above for resolution rules).

SKILL_DIR="${SKILL_DIR:-${CLAUDE_SKILL_DIR}}"
python3 "${SKILL_DIR}/scripts/check_prereqs.py"

If MISSING reported, see README.md for full setup instructions (install commands, API key setup, Remotion project init). The check is backend-aware: backend is resolved as TTS_BACKEND env var → user_prefs.json (global.tts.backend) → edge default, then only env vars required by that backend are validated.

Overview

Automated pipeline for professional Bilibili horizontal knowledge videos from a topic.

Target: Bilibili horizontal video (16:9)

• Resolution: 3840×2160 (4K) or 1920×1080 (1080p)
• Style: Clean white (default)

Tech stack: Coding agent + TTS backend + Remotion + FFmpeg

Output Specs

Execution Modes

Agent behavior: Detect user intent at workflow start:

• "Make a video about..." / no special instructions → Auto Mode
• "I want to control each step" / mentions interactive → Interactive Mode
• Default: Auto Mode

Auto Mode (Default)

Full pipeline with sensible defaults. Mandatory stop at Step 9:

1. Step 9: Launch Remotion Studio — user reviews in real-time, requests changes until satisfied
2. Step 10: Only triggered when user explicitly says "render 4K" / "render final version"

Users can override any default in their initial request:

• "make a video about AI, burn subtitles" → auto + subtitles on
• "use dark theme, AI thumbnails" → auto + dark + imagen
• "need screenshots" → auto + media collection enabled

Interactive Mode

Prompts at each decision point. Activated by:

• "interactive mode" / "I want to choose each option"
• User explicitly requests control

Technical Rules

Hard constraints for video production. Visual design remains the agent's creative freedom within these rules:

Additional Resources

Load these files on demand — do NOT load all at once:

• references/workflow-steps.md: Index of the 15-step workflow split across three phase files. Load at workflow start to locate which phase file to pull:
  • workflow-script.md — Pre-workflow + Startup + Steps 1-4 (scripting)
  • workflow-production.md — Steps 5-11 (media, TTS, Remotion, render, BGM)
  • workflow-publish.md — Steps 12-15 (subtitles, publish, cleanup, shorts)
• references/design-guide.md: Visual minimums, typography, layout patterns, checklists, animation safety rules. MUST load before Step 9.
• references/design-learning.md: Extracting visual patterns from reference videos/images, style profiles. Load only when the user provides a reference or manages profiles.
• references/azure-tts-pitfalls.md: Voice selection guide, Multilingual gotchas, SSML pitfalls, style support matrix. Load when choosing Azure voice/style or debugging hoarse/missing/glitchy audio.
• references/troubleshooting.md: Error fixes, BGM options, preference commands, preference learning. Load on error or user request.
• templates/presets/kinetic-typography/: Bold type-driven preset (black BG + mint/yellow accents + character-pop animations). Use for opinion / argument / declaration videos. Load README first.
• examples/: Real production video projects. The agent may reference these for composition structure and timing.json format.

Script suite dispatcher (optional)

All scripts in ${SKILL_DIR}/scripts/ are reachable through one hierarchical entry point:

python3 ${SKILL_DIR}/scripts/cli.py --help                 # list resources
python3 ${SKILL_DIR}/scripts/cli.py <resource> --help      # list actions
python3 ${SKILL_DIR}/scripts/cli.py <resource> <action> --help  # forwards to underlying script
python3 ${SKILL_DIR}/scripts/cli.py schema [<method>]      # JSON parameter schema

Routes (all forward to the existing scripts; direct invocations keep working): tts run|validate, verify, audit beats, shorts gen, design list|show|delete|add, prereqs, prefs get|migrate|backend|bgm-path, schema [<method>].

The dispatcher is optional — every snippet in this file uses the direct script path for back-compat. Agents that prefer one entry point with discoverable help should prefer cli.py.

Pre-render audit (recommended)

Before launching Studio for Step 9 review:

python3 ${SKILL_DIR}/scripts/audit_beat_sync.py <Video.tsx> <timing.json>

Prints a beat-vs-narration alignment table and flags drift > 1.5s. Especially important for kinetic-typography style where each beat must hit a specific narration moment.

End-of-pipeline acceptance gate (MANDATORY)

After Step 13, before declaring the video done:

python3 ${SKILL_DIR}/scripts/verify_output.py videos/{name}/

Auto-fixes common omissions (creates final_video.mp4 if missing; disable with --no-fix), validates resolution/codec/duration, checks audio-timing drift, sanity-checks publish_info.md. Exit 0 = green light to publish; exit 2 = warnings only, still publishable. For machine-readable output add --format json (auto when piped). Full flag reference in references/workflow-publish.md.

Directory Structure

project-root/                           # Remotion project root
├── src/remotion/                       # Remotion source
│   ├── compositions/                   # Video composition definitions
│   ├── Root.tsx                        # Remotion entry
│   └── index.ts                        # Exports
│
├── public/                             # Remotion default (unused — use --public-dir videos/{name}/)
│
├── videos/{video-name}/                # Video project assets
│   ├── topic_definition.md             # Step 1
│   ├── topic_research.md               # Step 2
│   ├── podcast.txt                     # Step 4: narration script
│   ├── podcast_audio.wav               # Step 8: TTS audio
│   ├── podcast_audio.srt               # Step 8: subtitles
│   ├── timing.json                     # Step 8: timeline
│   ├── thumbnail_*.png                 # Step 7
│   ├── output.mp4                      # Step 10
│   ├── video_with_bgm.mp4             # Step 11
│   ├── final_video.mp4                 # Step 12: final output
│   └── bgm.mp3                         # Background music
│
└── remotion.config.ts

Important: Always use --public-dir and full output path for Remotion render:

npx remotion render src/remotion/index.ts CompositionId videos/{name}/output.mp4 --public-dir videos/{name}/

Naming Rules

Video name {video-name}: lowercase English, hyphen-separated (e.g., reference-manager-comparison)

Section name {section}: lowercase English, underscore-separated, matches [SECTION:xxx]

Thumbnail naming (16:9 AND 4:3 both required):

Public Directory

Use --public-dir videos/{name}/ for all Remotion commands. Each video's assets (timing.json, podcast_audio.wav, bgm.mp3) stay in its own directory — no copying to public/ needed. This enables parallel renders of different videos.

# All render/studio/still commands use --public-dir
npx remotion studio src/remotion/index.ts --public-dir videos/{name}/
npx remotion render src/remotion/index.ts CompositionId videos/{name}/output.mp4 --public-dir videos/{name}/ --video-bitrate 16M
npx remotion still src/remotion/index.ts Thumbnail16x9 videos/{name}/thumbnail.png --public-dir videos/{name}/

Workflow

Progress Tracking

At Step 1 start, use your agent's task tracker (Claude Code TaskCreate / Codex todo list / equivalent) to create one task per step. Mark in_progress on start, completed on finish. Files in videos/{name}/ (e.g. podcast.txt, timing.json, output.mp4) act as the durable record of what completed — if a session is interrupted, inspect the directory to determine where to resume.

 1. Define topic direction → topic_definition.md
 2. Research topic → topic_research.md
 3. Design video sections (5-7 chapters)
 4. Write narration script → podcast.txt
 4.5. Pronunciation pre-flight (zh-CN only) → videos/{name}/phonemes.json
 5. Collect media assets → media_manifest.json
 6. Generate publish info (Part 1) → publish_info.md
 7. Generate thumbnails (16:9 + 4:3) → thumbnail_*.png
 8. Generate TTS audio → podcast_audio.wav, timing.json
 9. Create Remotion composition + Studio preview (mandatory stop)
10. Render 4K video (only on user request) → output.mp4
11. Mix background music → video_with_bgm.mp4
12. Add subtitles (optional) → final_video.mp4
13. Complete publish info (Part 2) → chapter timestamps
14. Verify output & cleanup
15. Generate vertical shorts (optional) → shorts/

Validation Checkpoints

After Step 8 (TTS):

• podcast_audio.wav exists and plays correctly
• timing.json has all sections with correct timestamps
• podcast_audio.srt encoding is UTF-8

After Step 10 (Render):

• output.mp4 resolution is 3840x2160
• Audio-video sync verified
• No black frames

Key Commands Reference

See CLAUDE.md for the full command reference (TTS, Remotion, FFmpeg, shorts generation).

User Preference System

Skill learns and applies preferences automatically. See references/troubleshooting.md for commands and learning details.

Storage Files

Priority

Final = merge(Root.tsx defaults < global < topic_patterns[type] < current instructions)

User Commands

Troubleshooting & Preferences

Full reference: Read references/troubleshooting.md on errors, preference questions, or BGM options.

Video Podcast Maker

中文文档

Automated pipeline to create professional video podcasts from a topic. Supports Bilibili, YouTube, Xiaohongshu, Douyin, and WeChat Channels with multi-language output (zh-CN, en-US). Combines research, script generation, multi-engine TTS (Edge/Azure/Doubao/CosyVoice), Remotion video rendering, and FFmpeg audio mixing.

Works with: Claude Code · OpenClaw (ClawHub) · OpenCode · Codex — any coding agent that supports SKILL.md

Publish to: Bilibili · YouTube · Xiaohongshu · Douyin · WeChat Channels

No coding required! Just describe your topic in plain language — the coding agent guides you through each step interactively. You make creative decisions, the agent handles all the technical details. Creating your first video podcast is easier than you think.

Note: This project is still under active development and may not be fully mature yet. We are continuously iterating and improving. Your feedback and suggestions are greatly appreciated — feel free to open an issue or reach out!

Features

• Topic Research - Web search and content gathering
• Script Writing - Structured narration with section markers
• Multi-TTS - Edge TTS (free), Azure Speech, Volcengine Doubao, CosyVoice, ElevenLabs, Google Cloud TTS, OpenAI TTS
• Remotion Video - React-based video composition with animations
• Visual Style Editing - Adjust colors, fonts, and layout in Remotion Studio UI
• Real-time Preview - Remotion Studio for instant debugging before render
• Auto Timing - Audio-video sync via timing.json
• BGM Mixing - Background music overlay with FFmpeg
• Subtitle Burning - Optional SRT subtitle embedding
• 4K Output - 3840x2160 resolution for crisp uploads
• Chapter Progress Bar - Visual timeline showing current section during playback
• Bilingual TTS - Chinese/English mixed narration with Azure Speech or CosyVoice
• Pronunciation Correction - Global + per-project phoneme dictionaries for Chinese polyphone fixes
• Bilibili Templates - Ready-to-use Remotion templates (Video.tsx, Root.tsx, Thumbnail.tsx, podcast.txt) for quick project scaffolding
• Component Library - Reusable visual building blocks (ComparisonCard, Timeline, CodeBlock, QuoteBlock, FeatureGrid, DataBar, StatCounter, FlowChart, IconCard, DiagramReveal, AudioWaveform, LottieAnimation, MediaSection, SectionLayouts, AnimatedBackground) for composing rich section layouts
• Preference Learning - Auto-learns user style preferences (colors, fonts, speech rate) and applies them to future videos
• Multi-Platform - Bilibili, YouTube, Xiaohongshu, Douyin, and WeChat Channels with independent platform and language settings
• Multi-Language - Chinese (zh-CN) and English (en-US) script templates, TTS voices, subtitle fonts
• Subtitle Preferences - Custom font, size, color, outline; toggle subtitle burning on/off
• Configurable CTA - Auto (Bilibili triple/YouTube subscribe), animation, text, or custom

Platform Optimizations

Bilibili:

• Script Structure - Welcome intro + call-to-action outro (一键三连)
• Chapter Timestamps - Auto-generated MM:SS format for B站 chapters
• Thumbnail Generation - AI (imagen/imagenty) or Remotion, auto-generates 16:9 + 4:3 versions
• Visual Style - Bold text, minimal whitespace, high information density
• Publish Info - Title formulas, tag strategies, description templates

YouTube:

• SEO Optimization - Title <70 chars, keyword-rich description, tags and hashtags
• Chapters - Auto-generated YouTube chapter timestamps (first line at 0:00)
• CTA - "Like, Subscribe & Share" text animation or custom

Xiaohongshu (小红书):

• Title - Max 20 chars, punchy and emoji-friendly
• Description - 200-500 chars, 种草/knowledge-sharing style with emoji
• Hashtags - #话题# format (double hash), 5-10 tags
• Thumbnail - 3:4 (1080x1440) for feed optimization
• CTA - "点赞收藏加关注" text animation

Douyin (抖音):

• Format - Vertical shorts only (9:16), no horizontal long-form
• Description - 100-200 chars, casual and conversational with emoji
• Hashtags - #话题 format (single hash), 3-8 tags
• CTA - "点赞关注" text only (no animation)

WeChat Channels (微信视频号):

• Format - Vertical shorts only (9:16), no horizontal long-form
• Description - 100-300 chars, knowledge-sharing style for forwarding
• Hashtags - #话题 format (single hash), 3-8 tags
• CTA - "点赞关注，转发给朋友" text only (no animation)

Workflow

⚠️ For the human reading this (not the AI): manually polish podcast.txt, repeatedly

This section is for you, the human — not the agent. Every downstream step — TTS narration, subtitles, section transitions, animation timing, final cut — is derived from this single podcast.txt. A weak script renders into 4K garbage. No amount of polish downstream saves it.

The AI-generated draft is a starting point, nothing more. Do these yourself — don't hand them off to the AI:

1. Mentally read it as the narrator. Treat each sentence as one breath — if a line forces you to "catch your breath" or backtrack to parse, fix it. Where you stumble silently is where TTS stumbles audibly.
2. Revise at least three times.
   • Pass 1: typos, awkward phrasing, tongue-twisters
   • Pass 2: cut filler, cut throat-clearing intros ("So today we're going to talk about…"), cut redundancy
   • Pass 3: tune rhythm — where to pause, where to break a long sentence, which word carries the stress
3. Read each [SECTION:xxx] block end-to-end. Confirm each section opens with a hook and lands a clean transition into the next — not a bullet-point dump.
4. Audit numbers, proper nouns, and English terms separately. ~90% of TTS mispronunciations live here. If pronunciation is wrong, add it to phonemes.json; if it just sounds awkward, rewrite it.
5. Know your length budget. Estimate ~280 zh-CN chars/min or ~150 en words/min. A 5–10 min video means ~1400–2800 chars / 750–1500 words. Don't pad to fill time.

The only acceptance test: read through it once in your head — does any line make you wince? If yes, don't move on to Step 8 (TTS) yet. Otherwise you're just rendering 4K of something even you don't want to hear.

Related Skills

This skill depends on remotion-best-practices and works alongside other optional skills:

• remotion-best-practices - Official Remotion best practices (required, provides core Remotion patterns and guidelines)
• find-skills - Official skill discovery tool (optional, helps find and install additional skills)
• ffmpeg - Advanced audio/video processing (optional)
• imagen / imagenty - AI thumbnail generation (optional)

Requirements

System Requirements

Installation

# macOS
brew install ffmpeg node python3

# Ubuntu/Debian
sudo apt install ffmpeg nodejs python3 python3-pip

# Python dependencies
pip install azure-cognitiveservices-speech dashscope edge-tts requests

Project Setup (Required)

Important: This skill requires a Remotion project as the foundation.

Understanding the components:

# Step 1: Create a new Remotion project (base framework)
npx create-video@latest my-video-project
cd my-video-project
npm i  # Install Remotion dependencies

# Step 2: Verify installation
npx remotion studio  # Should open browser preview

If you already have a Remotion project:

cd your-existing-project
npm install remotion @remotion/cli @remotion/player zod

API Keys Required

Environment Variables

Add to ~/.zshrc or ~/.bashrc:

# TTS Backend: edge (default, free), azure, doubao, cosyvoice, elevenlabs, google, openai
export TTS_BACKEND="edge"                            # Default (free), or "azure" / "doubao" / "cosyvoice" / "elevenlabs" / "google" / "openai"

# Azure TTS (high quality)
export AZURE_SPEECH_KEY="your-azure-speech-key"
export AZURE_SPEECH_REGION="eastasia"

# Volcengine Doubao TTS (alternative backend)
export VOLCENGINE_APPID="your-volcen