agentleFS
Sign inStart free
glifxyzglifxyz/glif-mcp-server/skills/make-images/SKILL.md

make-images

Skill213 starsChanged 37 days agoMITView on GitHub

Make new still images from a brief with Glif and save them to a folder: concept art, a series of variations, mockups, posters, illustrations, wallpapers, reference boards or any picture that isn't tied to a web page, app store or game engine. Use when: "make an image of", "generate a picture", "draw me", "I need some concept art", "make 10 variations", "one in each of these styles", "mockup of", "poster for", "wallpaper", "fake screenshot of", "what would X look like as Y", "mood board images", "images for my deck or doc". NOT for: editing an existing image (use edit-images), images that go into a site you're building (use web-visuals), logos or icons (use app-icons-and-logos), store or ad graphics (use app-promo-materials), game art (use game-assets), video or audio (use audio-and-video).

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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
  1. Distinctive:

    Use real people's faces only with their consent.

    line 23
  2. Very specific:

    Use the media's MIME type for the extension.

    line 41
  3. Distinctive:

    Check the pixel size. Ask Glif to redo a wrong one in the same project.

    line 42
  4. Distinctive:

    Check any text letter by letter.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Make the images the user described and put them where they can find them. The user's explicit instructions come first, within host safety rules.

1. Write the brief

Say what each image shows, not which model to use. Include:

  • Subject and scene: what is in the picture and what it is doing.
  • Style: a named look (pixel art, watercolor, 1990s desktop screenshot), a palette, lighting and mood. Name a reference style, not a living artist.
  • Size: aspect ratio and pixel size, such as 16:9 at 1920×1080 or 1:1 at 1024×1024.
  • Count: how many images, and what varies between them.
  • Text in the image: the exact words in quotes, kept short. Say "no text" otherwise.
  • References: upload a logo, sketch or style sample with upload_file, only after the user agrees to send that file to Glif. With shell access, pass filename (and kind if unclear), then run the returned command with the real path. Pass the returned fileUrl in attachments. Say what to keep from it: "Use the attached mark exactly, don't redraw it."

A series goes in one brief. Number the items and give each a short name and what makes it different, so Glif can title the outputs to match.

Use real people's faces only with their consent. Don't make images that pass as real photos of real people or events.

2. Check cost, then run once

Generation spends the user's Glif credits. A direct request ("make an image of…") is approval. If the images were your own idea, list them in a line each and wait for a yes.

Call compose_project once with the whole brief. Share the projectUrl. Follow the run with get_job_status, passing the jobId as job_id. A series can take minutes, so keep working meanwhile. A pending status or timeout is not a reason to start another run, and never retry a failure on your own.

3. Save and check

Download every result. In a codebase, use a folder the build does not ship, with a subfolder per series. Check the ignore and build config first, and never use public/ or static/. Elsewhere, use the folder the user named or the current directory.

curl -fsSL "<media url>" -o drafts/posters/01-noir.png
file drafts/posters/01-noir.png
  • Name files after the series item: 01-noir.png, 02-pastel.png.
  • Use the media's MIME type for the extension. Don't overwrite an existing file; add -v2.
  • Check the pixel size. Ask Glif to redo a wrong one in the same project; never upscale locally.
  • Check any text letter by letter.
  • For more than 6 images, build a contact sheet so the user can scan them in one look.

4. Report

Say what you made, where the files are and the projectUrl. Point out the ones that missed the brief. To change one image or add to the series, continue the same project with its project_id and name the item.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free