Claude + Gemini Video Understanding Skill
A free custom Claude skill that connects Claude Code to Gemini Video Understanding API so Claude can watch, summarize, and batch-rename video clips.
Before & after
Same prompt, same agent — the only difference is the skill.
Video organization is definitely helpful when you have a lot of clips! You'd probably want to watch through them, figure out what each one is about, and then rename them with a naming convention that makes sense—maybe by date, topic, or scene. Some tools can help with this, but honestly a lot of people just do it manually if they have the time. You could organize them by folder too, like by day or by scene. Tags or labels might help you find things later. It depends on how organized you want to be and how much time you have. Do you have a naming scheme already, or are you starting from scratch? And are they organized in any way right now?
VIDEO BATCH ORGANIZATION (20 clips, automated summarization + renaming) STEP 1: Upload folder to processing queue SYSTEM: Claude + Gemini Video Understanding API pipeline JOB: Analyze 20 clips in fan-out workflow (4 sub-agents, 5 clips each) STEP 2: Summarization results (returned per clip) CLIP 001: "Host explains product feature, B-roll of tool in use, 2 min 14 sec" CLIP 002: "Customer testimonial, speaks to transformation, 1 min 47 sec" CLIP 003: "B-roll only: office environment, typing, screen recordings, 45 sec" CLIP 004: "Blooper: guest laughs, awkward pause, 30 sec" CLIP 005: "Host walking, talking to camera, energy high, 1 min 22 sec" CLIP 006 through CLIP 020: Analyzed and summarized (landscape shots, product demos, audio-only segments, transitions) STEP 3: Auto-renamed files (based on content) feature-explanation-2-14.mp4 customer-testimonial-147.mp4 broll-office-setup-45.mp4 blooper-laughter-30.mp4 host-talking-energy-82.mp4 STEP 4: Organized by category (auto-tagged) /primary-content (clips for main video): 8 clips /broll (supporting background): 4 clips /testimonials (customer stories): 3 clips /outtakes (bloopers, unusable): 2 clips /review-later (unclear purpose): 3 clips TIME SAVED: 2 hours (manual watching + naming) → 15 minutes (automation + light review) OUTPUT READY: All files searchable by name + category. Drop into editing software immediately.
About this skill
name: claude-gemini-video-understanding description: Use when A free custom Claude skill that connects Claude Code to Gemini Video Understanding API so Claude can watch, summarize, and batch-rename video clips.
Claude + Gemini Video Understanding Skill
Built with Skill Creator, this skill wires Claude Code to Gemini Video Understanding API: Claude is the hands, Gemini is the eyes. Summarizes and renames folders of clips in a fan-out workflow with sub-agents. Free download: free.erictech.ca/claude-gemini-video
What you get
- A working Claude-to-Gemini video-analysis pipeline for summarizing/renaming clips.
Customize your output
- Adjust the summarization prompt or renaming convention.
Example output
Batch-summarized and renamed a folder of 17 clips in one run.
Best for
Video editors/creators wanting AI-assisted clip organization.
SKILL.md preview
---
name: claude-gemini-video-understanding
description: Use this skill when Claude needs to watch, summarize, and batch-rename folders of video clips using the Gemini Video Understanding API.
version: 1.0.0
category: Development / Media
author: AgentVolt
license: proprietary
tags:
- development
- media
- standard
---
# Claude + Gemini Video Understanding Skill
Wires Claude Code to the Gemini Video Understanding API so Claude can watch, summarize, and batch-rename folders of video clips in a fan-out workflow.
## When to use
… (sign up to view the full skill)More video & editing skills
View all Video & Editing skills →Matts Peeker
A Claude skill that watches any video (not just its transcript) to capture visual context Claude would otherwise miss.
Claude Video Skill
An open-source Claude skill that watches and analyzes videos (YouTube, TikTok, Zoom, Loom) frame by frame, not just via transcript.
Mograph Builder
Motion graphics from an AI video model: one style-locked motion sheet, then transcript-timed prompt packs for collage, type and B-roll.
Remotion Builder
Motion graphics as code: a script becomes deterministic Remotion scenes with charts, counters, maps and particle worlds, frame-exact.