agentleFS
pradeep-sankaran-07pradeep-sankaran-07/team-os-gdrive-template/.agents/skills/create-branded-image/SKILL.md

create-branded-image

Skill0 starsChanged 2 months agoMITView on GitHub

Generate original, on-brand Acme images. Use whenever someone wants a picture made rather than written: a marketing or campaign visual, a blog featured image, a social or ad image, a banner, a hero, an event or booth graphic, an illustration, a placeholder or mockup. Triggers: "create an image", "generate a visual/graphic", "make a picture of...", "I need a hero image for...", "a featured image for the blog", "a banner for the event". This applies BY DEFAULT to every image made for the company, the same way acme-brand applies to every document. Nobody has to ask for it. The only opt-out is an explicit "unbranded" or "in the customer's branding". NOT for charts or data visualisations, NOT for diagrams and flowcharts, NOT for product UI screens, and NOT for laying out a document (use acme-brand).

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. Very specific:

    Read references/house-style.md and compose the prompt from it.

    line 42
  2. Distinctive:

    Never render a real company's logo or trademark, including Acme's own.

    line 61
  3. Distinctive:

    Never generate an image of an identifiable real person.

    line 63
  4. Distinctive:

    Never present a generated image as a photograph of an actual customer, facility or event.

    line 64

≈ “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.

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

Create a branded image

This skill is mandatory by output type. Assume every image represents the company unless told otherwise. The opt-out is explicit words like "unbranded" or "in the customer's branding", never your own judgement that this one probably does not matter.

Step 1 - Establish the brief

What is the image for, where will it appear, and what aspect ratio does that placement need? A hero, a social card and a booth panel are three different images, and producing one of them for all three placements wastes the run.

Step 2 - Build the prompt from the house style

Read references/house-style.md and compose the prompt from it. Do not hand-roll a prompt and do not fall back to a generic look: the house style is the whole reason this skill exists rather than a raw image call.

Take colour from acme-brand's token file, by value, at prompt time.

Step 3 - Generate and file

Produce the requested number of variants. File them to output_location with a one-line README.md in the folder saying what the brief was, so a variant found six months later is identifiable.

Step 4 - If it will be published

acme-marketing runs alongside this skill. This one makes the picture; it does not decide what the picture is allowed to claim or imply.

Non-negotiables

  • Never render a real company's logo or trademark, including Acme's own. The logo is composited afterwards from the brand assets, never generated.
  • Never generate an image of an identifiable real person.
  • Never present a generated image as a photograph of an actual customer, facility or event.

Things that have gone wrong before

  • Generated logos are always wrong. Why: a model reproduces the shape of a mark without its geometry, and the result passes a glance and fails a brand review. Composite the real asset instead. (2026-09-16)

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.