Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Video Recap (End-to-End Narrated Recap Generator)

An orchestration skill that turns a raw video file into a finished narrated recap: understanding, story planning, cutting, AI voiceover and assembly.

Image & VideoAdvanced★ 470⑂ 86AI score 9/10Last updated: Aug 24, 2026

What it does

  • Takes .mp4/.mov/.mkv/.webm input, analyzes scenes, speech and characters, then forces the agent to plan in a real editorial order: directorial intent → story beats → shot selection → narration copy.
  • Acts as a lightweight orchestrator over video-understanding → script planning → video-cut → video-voiceover → video-assemble. Stages talk only through JSON/MP4 artifacts in work_dir, so a run is resumable by re-issuing the same command.
  • Three edit modes: full (full narration), cut (condensed recap, supports multiple source videos), dub (translate an English video and replace speech with a cloned voice of the original speaker).
  • Produces SRT/ASS subtitles, supports measuring the original subtitle band (tools/measure_subtitle.py), CapCut/剪映 project export, and optional advisory MiMo QC reports.
  • Hard timeline constraints: source audio may only be cut at reliable sentence/silence boundaries, and narration must use whole per-segment audio — preventing clipped, half-spoken lines.

Who it's for

  • Creators and editors producing recurring recap videos for dramas, films, documentaries or explainer content.
  • Channel owners who need to condense several episodes into one ~10-minute summary.
  • Localization teams who want translated dubs that keep the original speaker's timbre.
  • Requires ffmpeg plus a MiMo API key (Fish Audio optionally for TTS); output narration is Chinese-first by default.

Examples

  1. Single video: python3 scripts/recap.py ep1.mp4 --work-dir wd --context "show background" → the script writes a creative brief and pauses → you author recap_story_plan.json and narration.json → rerun the same command to render recap_ep1.mp4.
  2. Multi-episode digest: python3 scripts/recap.py ep1.mp4 ep2.mp4 --edit-mode cut --target-duration 10m.
  3. Voice-cloned dubbing: python3 scripts/recap.py talk.mp4 --edit-mode dub, supply per-sentence translations on the original timings, rerun to get dub_talk.mp4.
  4. Extra review: add --mimo-qc both for advisory pre-assemble and post-render QC notes in mimo_qc.json.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/zenstory-ai/video-recap-skills/HEAD/skills/video-recap/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/video-recap folder from the GitHub repo zenstory-ai/video-recap-skills into my ~/.claude/skills/zenstory-ai-video-recap/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/zenstory-ai/video-recap-skills.git /tmp/video-recap-skills && mkdir -p ~/.claude/skills && cp -r /tmp/video-recap-skills/skills/* ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Prerequisites: verify python3 --version and ffmpeg -version. Install ffmpeg if missing: macOS brew install ffmpeg, Ubuntu sudo apt install ffmpeg, Windows choco install ffmpeg.
  2. Clone the repo: git clone https://github.com/zenstory-ai/video-recap-skills.git
  3. Install the skills: mkdir -p ~/.claude/skills && cp -r video-recap-skills/skills/* ~/.claude/skills/ — copy the sibling skills (video-understanding, video-cut, video-voiceover, video-assemble) too, not just video-recap.
  4. Set credentials: export MIMO_API_KEY=your_key (add it to ~/.zshrc / ~/.bashrc to persist). For Fish Audio TTS also set export FISH_API_KEY=....
  5. Self-check: python3 ~/.claude/skills/video-recap/scripts/recap.py --doctor.
  6. Restart Claude Code, then ask something like "generate a narrated recap from this video file".
  7. The first run intentionally stops after the creative brief. Once the plan/narration JSON is written, rerun the exact same command to continue through rendering.
View source on GitHub ↗License: MIT