Matts Peeker
A Claude skill that watches any video (not just its transcript) to capture visual context Claude would otherwise miss.
Before & after
Same prompt, same agent — the only difference is the skill.
I can read the transcript of the video for you, which will give you the main points and what's being discussed. From the transcript, I can tell you what topics are covered and the key takeaways. However, I won't be able to see the actual visuals or any graphics that are on screen during the video. If the visual content is important to your analysis, you might want to describe what you see or watch the video yourself. Let me know what you'd like to extract from the transcript.
VISUAL ANALYSIS: VIDEO FRAME-BY-FRAME FRAME 0:00-0:15 • Speaker: seated, casual blue shirt, laptop visible on desk behind • Graphics: 3-column comparison chart appears lower-third (no audio mention) • On-screen text: "Agent Performance Q3 2026" (title + date) • Inference: Visual teases upcoming comparison not yet spoken FRAME 0:45-1:30 • Screen share: spreadsheet with 12 rows, highlighting columns B-D in yellow • Highlight: speaker points to row 7, pauses 3 seconds (emphasizing that data point) • Visual context: spreadsheet shows name column, 3 metric columns; audio talks metrics generically • Missing from audio: actual row labels not mentioned verbally FRAME 2:10-2:45 • Gesture: speaker taps screen 4x, each tap on different chart quadrant • Reaction: nods, small smile after quadrant 3 (nonverbal confidence signal) • Chart: pie chart with 6 slices; slice 1 = 38%, slice 2 = 22%, others smaller • Note: audio says "majority of users," visual shows 38%, confirming statement KEY VISUAL INSIGHTS • Nonverbal cues suggest confidence in Q3 results • On-screen breakdowns precede verbal explanation (visual-first pacing) • Pie chart granularity (6 segments) not mentioned in script; visual adds detail
About this skill
name: matts-peeker description: Use when A Claude skill that watches any video (not just its transcript) to capture visual context Claude would otherwise miss.
Matts Peeker
Most AI video analysis just reads the transcript. Matts Peeker actually watches the video frame content so Claude can reference what is shown on screen, not just what is said.
What you get
- Frame-level video understanding for Claude, beyond transcript-only analysis.
Customize your output
- N/A
Example output
Claude describing on-screen visuals and events a transcript alone would miss.
Best for
Creators and researchers who need Claude to understand video content, not just audio.
SKILL.md preview
---
name: matts-peeker
description: Use this skill when the agent needs to analyze what's actually shown on screen in a video, not just its transcript.
version: 1.0.0
category: Development / Media
author: AgentVolt
license: proprietary
tags:
- development
- media
- standard
---
# Matts Peeker
Watches a video's actual frame content so the agent can reference what's visually on screen — a slide, a UI demo, a diagram — instead of relying only on the spoken transcript.
## When to use
… (sign up to view the full skill)More video & editing skills
View all Video & Editing skills →Claude Video Skill
An open-source Claude skill that watches and analyzes videos (YouTube, TikTok, Zoom, Loom) frame by frame, not just via transcript.
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.
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.