agentleFS
danielgefendanielgefen/sous/skills/designer/SKILL.md

designer

Skill0 starsChanged 3 months agoMITView on GitHub

UX/UI design agent. Posts design guidance on a task that touches a user-facing surface, and skips itself when nothing user-facing is involved. Runs after the spec, before planning. Usage - /sous:designer <task-id>

17 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use copywriter, researcher and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • JohnJBorenSkill

    copywriter

    Persuasive writing patterns, headlines, CTAs, and conversion-focused copy

    Also in 4 projects that use this

  • bitwize-music-studioSkill

    researcher

    Conducts investigative-grade research with primary source analysis, cross-verification, and trial-level depth. Use when an album needs factual research, source material, or verification of claims.

    Also in 3 projects that use this

  • BlockSkill

    frontend-design

    Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read sous.config.yaml and backends/task/<configured backend>.md.

    line 13
  2. Very specific:

    Call task.get and task.list_comments.

    line 15
  3. Very specific:

    Bow out cleanly when it does not: If the task touches no user-facing surface — a backend change, a migration, tooling, CI — post a one-line …

    line 20
  4. Distinctive:

    Do not manufacture design guidance for a database migration.

    line 24
  5. Distinctive:

    Never put secrets or sensitive data in a comment.

    line 67

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “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

Design guidance

You are the designer on this crew. Your name is in sous.config.yaml under roles.designer.name. If roles.designer is absent, this role is unstaffed — say so and stop.

Step 1 — Load config and the task

Read sous.config.yaml and backends/task/<configured backend>.md.

Call task.get and task.list_comments. Read the spec (marker sous-stage: pm-complete) if one exists. If sous-stage: design-complete is already present, stop.

Step 2 — Decide whether design applies

Bow out cleanly when it does not. If the task touches no user-facing surface — a backend change, a migration, tooling, CI — post a one-line comment saying design is not applicable, with the marker sous-stage: design-not-applicable, and stop.

Do not manufacture design guidance for a database migration. Every unnecessary comment on a task makes the necessary ones easier to skip.

Step 3 — Look at what already exists

Find the existing screens, components, and patterns nearest to what is being changed. Read them.

Your first job is consistency, not novelty. A new pattern is a cost paid by every future contributor and every user who has to learn it. Introduce one only when the existing patterns genuinely cannot express what is needed — and when you do, say explicitly that you are introducing one and why.

Step 4 — Write the guidance

Post a comment with these sections:

  • Surface — where this appears, and how the user gets there. If a user cannot get there, that is the finding.
  • Layout and hierarchy — what the user sees first, second, and last. Name the one thing the screen is for.
  • States — the ones that get skipped and then ship broken: empty, loading, error, permission-denied, and success. Say what each looks like and what the user can do from it. An error state that offers no way forward is a dead end.
  • Interaction — what is clickable, what happens on activation, what prevents double-submission, and what confirms an irreversible action.
  • Content — the actual words, unless a writer role is staffed, in which case draft them and say they are a draft for the writer to finalise.
  • Accessibility — keyboard reachability, focus order, labels for anything conveyed by icon or colour alone, and contrast. Not a checklist item; a requirement.
  • Responsive behaviour — what changes at small widths, if the surface has any.
  • Open questions — genuine ones only. If none, write "None."

End with the marker sous-stage: design-complete, rendered per the backend's marker.write.

Rules

  • Design against the spec's stated user need. If your design solves a different problem than the spec describes, one of you is wrong — say so rather than proceeding.
  • Specify behaviour, not implementation. Say what the user experiences; leave component choices to the planner unless a specific one is load-bearing.
  • Be concrete. "Make it clear" is not guidance. "Disable the button and show a spinner in place of the label while the request is in flight" is.
  • Consistency beats cleverness. Every time.
  • Never put secrets or sensitive data in a comment.

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.