youtube-profile-spec
Skill2 starsChanged 49 days agoMITView on GitHub
Specification for a YouTube profile picture — 1:1, cropped to a circle. Triggered when generating or reviewing a channel profile picture.
In this repository23 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- analyse-channel-packagingSkill
- analyse-thumbnailsSkill
- channel-analyticsSkill
- channel-identitySkill
- channel-kickoffSkill
- competitor-boardSkill
- competitor-ideationSkill
- competitor-mappingSkill
- competitor-scanSkill
- image-generation-coreSkill
- image-refinementSkill
- keywords-seoSkill
- likeness-preservationSkill
- memory-reviewSkill
- monetisationSkill
- packagingSkill
- script-writingSkill
- thumbnail-best-practicesSkill
- thumbnail-inspirationSkill
- transparencySkill
- youtube-banner-specSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 33
If the creator's face appears, load likeness-preservation: in an avatar the identity is all there is.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Profile picture specification
WHEN
Any profile picture generation or review.
THE RULE THAT DECIDES EVERYTHING
It is uploaded square, 1:1, but YouTube crops it to a circle. Anything near the corners is lost.
And it is seen small: 32x32 pixels in the comments. That is the real size it has to be recognisable at, not the one in your editor.
EXECUTION ORDER
- See the current one
python3 tools/view_channel_packaging.py --only avatar --mdand open it. Generate
python3 tools/generate_image.py --type profile_image --prompt "..." \ --ref face_photo.jpg:likeness --dry-run--type profile_imagealready asks for a centred subject, a simple background, no text and legibility at 32x32.If the creator's face appears, load
likeness-preservation: in an avatar the identity is all there is.- Verify at real size, which is what almost nobody does:
python3 tools/export_image.py --image PATH --size 32x32 --out /tmp/avatar32.pngand open that 32x32 file. If you cannot tell who it is at that size, it does not work, however good it looks at 800x800. - Export the final one:
python3 tools/export_image.py --image PATH --type profile_image
COMPOSITION
- Subject centred and large: the circle eats the corners.
- Simple, single-colour background, high contrast against the subject.
- No text and no fine detail. At 32x32 it is a smudge.
- Colour-consistent with the banner.
SOURCES
source: generated for the image; derived for the resizing.
OUTPUT
- The final profile picture, exported with
--type profile_image. - The 32x32 check: whether it is still recognisable at comment size.
- The model that generated it, named explicitly.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

