sxsphinxstack

Skills / Make a thumbnail

Make a thumbnail

Make a thumbnail that tells the truth and earns the click. Composition, text readable at small size, correct export. Use when they say "make me a thumbnail" or they are about to upload a video where thumbnails matter.

Use this skill. Nothing to install.

Open in ChatGPT Open in Claude prefilled and ready; just hit send
for Gemini, Copilot, or Cursor: paste it, then say "use the make a thumbnail skill"

Don't have an agent?  ·  Raw file: skills/make-a-thumbnail.md

Watch the first two minutes

This is how a session goes when thumbnail advice has already put you off thumbnails. Click through it.

(pastes the skill) i need a thumbnail but every tutorial is like SHOCKED FACE, RED ARROW, and that feels gross
Then we already agree. The skill's first rule: the thumbnail shows something that actually happens in the video, in the video's actual tone. Truthful and clickable are compatible; that is the craft. What's the video?
me fixing the gear shifting on my bike, filmed on my phone
Good subject. One subject is the whole composition rule: your hands on the derailleur, filling a third to half the frame. Scrub the video for a frame where the hands and the mechanism are sharp and well lit, and grab it.
every frame is kind of dark and blurry
Then re-stage it: same bike, same grip, shot as a photo facing a window. That is still true, it's the real repair moment with better light. What we won't do is a face you never made or an arrow pointing at nothing.
should it have text on it?
Only if it says something the image can't. If the repair genuinely cost nothing, "FIXED FOR $0" in three heavy words earns its place; repeating your title doesn't, because they share the screen. Then the real test: zoom out until the image is about 120 pixels wide. If you can't tell what it is, we simplify and test again.

Scripted example of a real session.

Isn't this just clickbait?

A thumbnail is a promise about what the video contains, and the skill's whole job is making that promise vivid without lying. No shocked face you never make, no arrows at things that are not in the video, no "you won't believe" framing over ordinary content. Ordinary content with an honest, legible, high-contrast thumbnail beats the same content behind a bait one, because a viewer who got what the thumbnail promised stays, and the platforms measure that. The craft is composition and contrast, and the skill teaches both on your own footage.

What you end up with

An exported image, tested at the size people will actually see it, plus the decisions behind it. Here is the spec from the bike-repair session above.

Example
THUMBNAIL SPEC — BIKE SHIFTING FIX
Tool: Canva free tier  ·  Source: re-staged phone photo, window light
COMPOSITION 1
Subject: two hands on the derailleur, filling the right half of the frame
Light: window at camera left; subject reads light against the dark garage wall
Contrast bumped in the editor instead of adding decoration
Text: "FIXED FOR $0", heavy bold, solid backing bar, upper left 2
True: the fix used a hex key and a barrel adjuster, no parts
SMALL TEST 3
At 120 px wide: subject recognizable, text readable after cutting five words to three
Compared next to a screenshot of the feed; holds its own without imitating anyone
EXPORT
1280x720 JPG · under 2 MB · saved next to the video, named to match
  1. One subject. Thumbnails with three subjects have zero. Everything else in the frame exists to make the one subject read.
  2. The text does a different job than the title. They appear on the same screen, so saying the same thing twice wastes the strongest four words you get.
  3. The small test is the gate. Thumbnails are seen at postage-stamp size on phones. Nothing ships until it survives 120 pixels.

Anatomy of a thumbnail that reads

The skill checks every draft against this line:

one subject, a third to half the frame · light against dark, or dark against light · 3–4 heavy words, only if the image can't say it · still legible at 120 px

Questions people actually ask

Is this free?

The skill is free and so are the tools it uses: Canva free tier, Figma, GIMP, or even Google Slides. Photos come from your own footage or a phone photo you shoot during the session.

I'm not a designer. Can I really make one?

Yes, because the method is narrow: one subject, filled frame, high contrast, few words. The small test tells you honestly whether it worked, and a second draft takes minutes.

Should my face be in it?

If your face is in the video, it is a strong candidate, because eyes read at any size. The expression has to be one you actually make in the video. A staged shocked face over calm content is exactly what the skill refuses.

Do TikTok, Reels, and Shorts need thumbnails too?

Vertical platforms take a cover frame from the video itself, so the job there is picking that frame deliberately at upload. The full 1280x720 thumbnail is for YouTube.

What if I can't choose between two ideas?

Make both; it is ten minutes. Compare them at small size and pick on legibility and honesty. If the platform offers thumbnail testing, YouTube does, let real viewers settle it.

Where to go from here

The thumbnail is one piece of the upload. The rest:

Publish to YouTube Edit a short Start a YouTube channel

No video yet? Make one worth a thumbnail:

Screen-recording tutorial Before/after short How-to guide
Curious? Read the full skill — the exact instructions your agent gets
---
name: make-a-thumbnail
category: media
description: Make a thumbnail that tells the truth and earns the click. Composition, text readable at small size, correct export. Use when they say "make me a thumbnail" or they are about to upload a video where thumbnails matter.
---

# make-a-thumbnail

Make a thumbnail with someone for a real video they made.
A thumbnail is a promise about what the video contains; the job is
to make that promise vivid without lying. The session ends with an
exported image, tested at the size people will actually see it.

## Ground rules

- The thumbnail must be true. It shows something that happens in
  the video, in the video's actual tone. No shocked-face they never
  make, no red arrows pointing at things that are not in the video,
  no "you won't believe" framing over ordinary content. Truthful
  and clickable are compatible; that is the craft.
- Free tools: Canva free tier, Figma, GIMP, or even Google Slides.
  Photos come from their own footage — a frame grab from the video
  or a photo they shoot right now on their phone.

## Composition

1. Pick the subject: one face, one object, or one moment from the
   video. One. Thumbnails with three subjects have zero.
2. Frame grab or reshoot: scrub the video for a frame where the
   subject is sharp and well-lit, or stage the same real moment for
   a phone photo with better light (face a window).
3. Fill the frame. The subject should take a third to a half of the
   image. Faces work because eyes read at any size — if their face
   is in the video, it is a strong candidate for the thumbnail.
4. Contrast carries it: subject light against dark background or
   dark against light. If the grab is flat, bump contrast and
   brightness in the editor rather than adding decoration.
5. Text only if it adds what the image cannot say: 3–4 words max,
   heavy bold font, high contrast (add a solid or shadowed backing
   if the background is busy). The text and the video title should
   not say the same thing — they share the same screen.

## The small test

Thumbnails are seen at postage-stamp size on phones. Before calling
it done:

- Zoom the canvas out until the image is about 120 pixels wide, or
  export and look at it in a file browser icon view.
- Can you tell what it is? Can you read the text? If not, bigger
  subject, fewer words, more contrast — repeat until it survives.
- Line it up next to a screenshot of the platform's feed. It should
  hold its own without imitating anyone specific.

If they are choosing between two ideas, make both — it is ten
minutes — and compare at small size. Pick on legibility and honesty,
then note which one ran; if the platform offers thumbnail testing
(YouTube does), let real viewers settle it. That is A/B thinking
without inventing anything.

## Done

- Check the destination's current official dimensions, aspect ratio, format,
  and file-size limit before export. For platforms that select a video frame,
  choose that frame deliberately during upload.
- The exported file is saved next to the video, named to match.
- Both of you have seen it at real size and agree it promises only
  what the video delivers.

Uploading next? The publish-to-youtube skill covers the rest of the
package: title, description, and what to watch after it goes live.