agentleFS
DankosikDankosik/brandfile/skills/brand-directions/SKILL.md

brand-directions

Skill0 starsChanged 7 days agoMITView on GitHub

Style directions. Use when a confirmed brand brief has elements marked open or evolve, and the person must choose how the brand looks.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the average in one line and set it aside.

    line 18
  2. Distinctive:

    Write at least six one-line ideas, each starting from a different fact: a founder's decision, the tension, the non-obvious trait, a …

    line 20
  3. Distinctive:

    Run the swap test: with the name removed, would it fit a direct competitor?

    line 21
  4. Distinctive:

    Keep the ones whose structures differ most: three by default, two when little is open.

    line 23
  5. Very specific:

    Write one standalone HTML file per direction, brand/work/directions/A.html and so on.

    line 27
  6. Very specific:

    When status moves on, hand off to next.

    line 51

≈ “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

Brand Directions

Show a few whole directions that differ in how the page is built, on the brand's real material, and let the person choose by seeing. Done when brand/work/selection.md records a settled choice for every open and evolve element.

Run node brand/tools/brand.mjs status (if brand/tools/ is missing, hand off to brandfile); if next names another skill, hand off to it. Read brand/work/brief.md, brand/work/selection.md if it exists, and brand/BRAND.md if a brand is accepted, since its values are what keep means. Read the craft notes before choosing color and type, and the Cyrillic font notes for a brand that writes in Cyrillic.

A missing fact goes back: add the question under "Open questions" in the brief, set confirmed: no, and hand off to brand-interview.

Ideas in words first

Your first idea is the category's average. Get past it in words, before any HTML:

  1. Write the average in one line and set it aside.
  2. Name the category codes from the brief's competitors and references: lead colors, type, layout, tone.
  3. Write at least six one-line ideas, each starting from a different fact: a founder's decision, the tension, the non-obvious trait, a customer's phrase, a reference.
  4. Run the swap test: with the name removed, would it fit a direct competitor? Drop those.
  5. Give each survivor a structure: what the first screen shows, how dense it is, what leads the hierarchy, which section patterns appear, and what changes with the task. Anchor each choice to a fact in the brief; banning today's fashion only moves it.
  6. Keep the ones whose structures differ most: three by default, two when little is open.

Build each

Write one standalone HTML file per direction, brand/work/directions/A.html and so on. Copy the sheet there as index.html and list them; it opens from disk.

  • Same real material in every direction: their name and copy, their first screen, and the screens behind the brief's tasks, so the person compares directions, not content. Numbers, reviews, and quotes are their own; mark a missing one as a gap.
  • Repaint test: swap the colors and fonts of two directions. If the pages would then match, rebuild one with another structure.
  • keep elements are identical everywhere; evolve ones start from the current value.
  • The logo is their file; you never draw one, even when asked. Without a file, set the name in type and label it "not a logo".
  • Voice comes from their own good texts. If voice is open and they have none, show one message in each direction's voice.
  • Rationale closes each direction: the facts and references behind it, the category code it keeps and the one it breaks, what it gives up, and which part is taste. It claims nothing about sales or recognition.

Check every text pair with node brand/tools/brand.mjs contrast "#a" "#b" and every font file with node brand/tools/brand.mjs glyphs <file> <language> "<name>". Fonts load from brand/work/directions/fonts/, or from a hosted link when there is a network. Open the sheet in a browser and look at it; if you cannot, say so when you hand it over.

Stop: they choose

Give the path brand/work/directions/index.html, opening it if you can. Ask which fits the brand, not which is nicest. They answer in words, often mixed: "B, but the type from C". Build the next sheet from that. Give your own view only when asked, with its reason, marked as yours. When their taste and the brief conflict, say so and let them settle it. On "you decide", choose, give the reason, and record it as an assumption.

After each round, write brand/work/selection.md, one row per element, named exactly as in the brief. Fill Choice only for what they have settled, since a filled Choice is final. In Why, keep their reason and what they turned down, with its reason. Include the values, font sources, and file paths the book will need: the next stage reads this file, not the sheet.

| Element | Choice | Why |
|---|---|---|
| Color | B: #0B5FFF on white, with #14171F text | "Looks like a working tool". Turned down A's orange: "too much like a shop" |
| Type | C: Golos Text | Assumption: the person said "you decide" |

When status moves on, hand off to next.

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.