agentleFS
tim-osterhustim-osterhus/millrace-skills/skills/frontend-product-builder/SKILL.md

frontend-product-builder

Skill0 starsChanged 7 months agoApache-2.0View on GitHub

Guardrail for building frontend product surfaces from real workflows, state, and browser-verified interaction instead of generic landing-page scaffolds.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. About 0 are common, 17 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈17

    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:

    If the prompt is a one-shot build or design request with no follow-up available, infer reasonable defaults and keep moving to a complete …

    line 40
  2. Distinctive:

    Keep labels, numbers, and controls in code.

    line 43
  3. Distinctive:

    Verify the real surface in a browser on desktop and mobile before handoff.

    line 44
  4. Distinctive:

    Ask a clarifying question only when the request is genuinely ambiguous or unsafe.

    line 49
  5. Distinctive:

    Do not turn a dashboard, tool, or app request into a marketing shell.

    line 53
  6. Distinctive:

    Keep text, labels, metrics, and controls in code.

    line 54
  7. Very specific:

    Use browser-local-qa for the smallest reliable verification surface.

    line 55
  8. Distinctive:

    Start from the product category and actual job to be done.

    line 65
  9. Distinctive:

    If context is missing, surface the assumption and continue rather than asking a narrowing question unless the request is genuinely …

    line 68
  10. Distinctive:

    Write down the core objects, states, and actions the user expects.

    line 78
  11. Distinctive:

    Start from those objects, not from a generic hero, feature grid, or card stack.

    line 79
  12. Distinctive:

    If the surface is operational, map prompt nouns to named regions first, then use the domain's native UI forms: tables, lists, filters, …

    line 80
  13. Distinctive:

    Use real sample data and states instead of fake metrics and social proof.

    line 81
  14. Distinctive:

    Check the screen for vague labels, weak alignment, overflow, missing controls, or "looks polished but does nothing" behavior.

    line 83
  15. Distinctive:

    Check that the first screen shows the real task, not a generic promo page.

    line 98
  16. Distinctive:

    Check that the controls, states, and copy are specific to the product.

    line 99
  17. Distinctive:

    Check that any visual flourish helps the task instead of hiding missing substance.

    line 100
  18. Distinctive:

    If the result still feels generic, re-check whether it accidentally became a landing page or reused default component-library tropes.

    line 104

≈ “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

Frontend Product Builder

Purpose

Help agents build frontend product surfaces from the real workflow, state model, and browser-verified interaction instead of generic landing-page scaffolds. This is a guardrail, not a frontend course or a browser QA manual.

Use When

  • the task is a new frontend app, tool, dashboard, game, site, or redesign
  • the first screen should be the real product, not a marketing wrapper
  • the surface needs real state, controls, or responsive browser checks

Do Not Use When

  • the request is only a small style tweak inside a fixed design system
  • the work is backend-only, copy-only, or already owned by a more specific skill
  • you only need generic frontend advice or a browser-selection guide

Quick Start

  1. Classify the surface first: app, tool, dashboard, game, site, or other product UI.
  2. Name the real job, primary objects, states, and actions before choosing layout or visuals. If the prompt is a one-shot build or design request with no follow-up available, infer reasonable defaults and keep moving to a complete first-pass answer.
  3. Decide whether concepting or imagegen is actually needed. Use it only when the surface needs concrete visual direction or non-textual product art.
  4. Build the working first screen and core workflow before polish.
  5. Keep labels, numbers, and controls in code; use existing assets or imagegen only for visuals code cannot convincingly supply.
  6. Verify the real surface in a browser on desktop and mobile before handoff.

Operating Constraints

  • Operational surfaces should read as usable product, not a promo page or hero section.
  • One-shot build and design prompts should be answered directly in one response; do not default to clarification when a reasonable assumption can resolve the gap.
  • Ask a clarifying question only when the request is genuinely ambiguous or unsafe.
  • Preserve expressive categories such as games, portfolios, editorial work, and art-directed product surfaces; do not flatten them into generic SaaS cards.
  • Make at least one concrete interaction or state change explicit: select, edit, filter, toggle, inspect, or step through a workflow.
  • Reject at least one generic trope and name the better alternative, such as replacing a hero/card stack with a table, sidebar, or inspector that shows the actual workflow.
  • Do not turn a dashboard, tool, or app request into a marketing shell.
  • Keep text, labels, metrics, and controls in code; use assets or imagegen for visuals that code cannot convincingly supply.
  • Use browser-local-qa for the smallest reliable verification surface; this skill only requires that the real page gets checked on desktop and mobile.

Inputs This Skill Expects

  • The product category and the user's real goal.
  • Any existing UI, screenshot, brand system, or design system.
  • The key entities, states, filters, and actions the surface must support.
  • Any hard constraints on layout, tone, density, responsiveness, or art direction.
  • Missing optional context is acceptable; turn it into stated assumptions instead of blocking on a narrowing question unless the request is genuinely ambiguous or unsafe.

Output Contract

  • Start from the product category and actual job to be done.
  • For apps, tools, dashboards, and games, build the real workflow first, not a promo wrapper.
  • For one-shot operational-surface prompts, answer in one response with the requested plan, named layout regions, at least one concrete interaction or state path, one rejected generic trope, and browser verification steps.
  • If context is missing, surface the assumption and continue rather than asking a narrowing question unless the request is genuinely ambiguous or unsafe.
  • For operational surfaces, translate at least three nouns from the prompt into named UI regions, controls, or state panels.
  • For operational surfaces, show one concrete inspect-or-edit path for a selected object, not just a comparison table or summary strip.
  • Choose domain-specific hierarchy, density, controls, and content.
  • Make at least one interaction, mode, or state explicit on the screen.
  • Explain at least one generic trope you rejected and the better product-specific alternative.
  • Preserve intentional expressive design when the category calls for it, but keep the surface usable and legible.

Procedure

  1. Identify whether the request is a landing page or a product surface, and whether it is a one-shot prompt with no follow-up available.
  2. Write down the core objects, states, and actions the user expects. If optional context is missing, state the assumption and continue unless the request is genuinely ambiguous or unsafe.
  3. Start from those objects, not from a generic hero, feature grid, or card stack.
  4. If the surface is operational, map prompt nouns to named regions first, then use the domain's native UI forms: tables, lists, filters, sidebars, drawers, timelines, or inspectors.
  5. Use real sample data and states instead of fake metrics and social proof.
  6. Include at least one inspect-or-edit path for a selected object.
  7. Check the screen for vague labels, weak alignment, overflow, missing controls, or "looks polished but does nothing" behavior.
  8. Tune density and whitespace to the product instead of defaulting to the same spacious demo layout every time.

Pitfalls And Gotchas

  • Turning every request into a SaaS dashboard.
  • Using gradients, blobs, glassmorphism, glow, or orbs as default decoration.
  • Inventing fake dashboards, fake charts, or fake testimonials to fill space.
  • Making every UI rounded, airy, and shallow until it no longer matches the product.
  • Fixing style while leaving the main workflow vague or inert.
  • Banning all visual ambition and breaking games, portfolios, editorial work, or art tools.

Progressive Disclosure

Start with the smallest useful reading of the request, then widen only as needed to support the actual workflow. If the surface is operational, favor clarity and task completion; if it is expressive, favor the art direction the category deserves. Keep the skill short enough that it steers decisions instead of becoming a generic design manifesto. When in doubt, preserve the product skeleton and make the design more specific, not more generic.

Verification Pattern

  • Check that the first screen shows the real task, not a generic promo page.
  • Check that the controls, states, and copy are specific to the product.
  • Check that any visual flourish helps the task instead of hiding missing substance.
  • For operational surfaces, verify at least one interaction path such as filter, select, toggle, edit, or inspect.
  • For expressive surfaces, verify that the chosen style is intentional and category-appropriate, not the default AI-generated look.
  • Confirm the browser check covered the real page on desktop and mobile before handoff.
  • If the result still feels generic, re-check whether it accidentally became a landing page or reused default component-library tropes.

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.