agentleFS
ytvee-devytvee-dev/webdev-agent-kit/skills/frontend-layout-implementer/SKILL.md

frontend-layout-implementer

Skill7 starsChanged 21 days agoApache-2.0View on GitHub

Implement approved React/Next.js design specs using existing CSS Modules, Redux, TanStack, and Axios conventions. Route uninspected designs to intake and resolve product decisions before coding. Excludes unrelated stacks and styling systems.

The file

Show its rules and text

We split this skill into the rules it gives an agent

33 rules. About 0 are common, 13 are distinctive, and 20 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈13

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 20

    Very specific

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

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

    When the user asks to keep iterating until measurable visual, lint, build, or route acceptance criteria pass, implementation must follow a …

    line 52
  2. Very specific:

    Read common/approved-patterns.md and relevant anti-pattern rules.

    line 73
  3. Very specific:

    Read common/css-modules-specificity-rules.md for CSS Modules changes.

    line 74
  4. Distinctive:

    Read conditional UX rules only when the touched surface needs them: forms, navigation, data visualization, icons, or mobile responsive …

    line 75
  5. Very specific:

    Read common/agent-loop-policy.md and common/verification-loop-rules.md when measurable iteration or bounded retry is required.

    line 76
  6. Distinctive:

    Read relevant project overlays when present: stack, architecture, styling, verification, state, data-fetching, build, and workspace …

    line 77
  7. Very specific:

    Read the Design Implementation Spec, Design Direction Contract, and supplied visual references.

    line 78
  8. Distinctive:

    Read affected source files, styles, components, routes, and configs.

    line 79
  9. Very specific:

    Read references/implementation-rules.md when present.

    line 80
  10. Very specific:

    Do not read generated dist/ during normal runtime.

    line 82
  11. Very specific:

    When shared UI or compatible component wrappers are in scope, read common/frontend-design-system-rules.md and …

    line 84
  12. Very specific:

    When React/Next work changes loading, imports, subscriptions or expensive rendering, read the relevant section of …

    line 89
  13. Very specific:

    Use Project Context MCP when available.

    line 94
  14. Distinctive:

    Use Design Spec MCP when available.

    line 95
  15. Distinctive:

    Use Visual Reference MCP when available.

    line 96
  16. Very specific:

    Use context7 for React, Next.js, Redux, TanStack, Axios, TypeScript, or build-tool docs when implementation depends on current behavior.

    line 97
  17. Very specific:

    Use mdn for current HTML, CSS, Web API, accessibility, and compatibility facts.

    line 99
  18. Distinctive:

    Use rendered visual QA only when screenshot comparison, viewport evidence, overflow checks, or visible state verification are in scope.

    line 100
  19. Distinctive:

    Use Visual Diff MCP when available during final visual comparison.

    line 101
  20. Distinctive:

    If a named MCP is unavailable, report the missing capability before using a lower-confidence fallback.

    line 103
  21. Very specific:

    Check component coverage and the spec's product decision register before editing.

    line 110
  22. Very specific:

    Use frontend-architecture-planner before editing when route, state, data, form, build, workspace, or shared component ownership is unclear …

    line 116
  23. Very specific:

    Use loop-workflow-planner before editing when the task requires bounded retry, independent review, loop memory, or measurable iteration …

    line 117
  24. Distinctive:

    Keep UI state local unless existing React, Redux, or TanStack ownership requires otherwise.

    line 123
  25. Distinctive:

    Keep styles in CSS Modules by default.

    line 124
  26. Distinctive:

    Do not add dependencies, global tokens, UI libraries, or new styling systems without explicit approval.

    line 126
  27. Very specific:

    Use frontend-linter-manager after code-changing work when a lint command exists.

    line 128
  28. Very specific:

    Run relevant project verification commands from project/verification-profile.md.

    line 129
  29. Very specific:

    Run frontend-visual-qa only when the change needs rendered screenshot comparison, viewport evidence, overflow verification, or visible …

    line 130
  30. Very specific:

    Use frontend-quality-reviewer before final reporting when the user asks for review, the implementation has significant surface area, a loop …

    line 131
  31. Very specific:

    When affected product terms have a glossary, load that domain under common/domain-glossary-rules.md.

    line 151
  32. Distinctive:

    Keep visual QA routed separately.

    line 161
  33. Distinctive:

    No new package, styling system, global token, architecture layer, or UI library may appear without approval.

    line 173

≈ “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 Layout Implementer

Purpose

Implement a Design Implementation Spec or Design Direction Contract in a React or Next.js project while respecting existing CSS Modules, Redux, TanStack, Axios, architecture, styling, decomposition, UX gates, and verification conventions.

When the user asks to keep iterating until measurable visual, lint, build, or route acceptance criteria pass, implementation must follow a Loop Workflow Contract or hand off to loop-workflow-planner before editing.

When To Use

  • The user asks to build a screen, section, component, or static page from a Design Implementation Spec.
  • The user supplies screenshot-derived design material and asks for code in the current supported frontend project.
  • The implementation target is within React, Next.js, CSS Modules, Redux, TanStack, or Axios scope.

When Not To Use

  • The design intent has not been converted into a spec from supplied links or screenshots. Use design-screenshot-spec first, then resume requested code work.
  • The user asks only for live design inspection, canvas edits, or whiteboard work.
  • The project is outside the supported target stack unless the user explicitly changes scope.
  • The task is unrelated to frontend rendering or layout.

Required Context

  1. Read AGENTS.md.
  2. Read common/target-stack-policy.md.
  3. Confirm the classified task is frontend-layout, feature/development, bugfix, refactor, or optimization for a rendered React or Next.js surface.
  4. Read common/approved-patterns.md and relevant anti-pattern rules.
  5. Read common/css-modules-specificity-rules.md for CSS Modules changes.
  6. Read conditional UX rules only when the touched surface needs them: forms, navigation, data visualization, icons, or mobile responsive behavior.
  7. Read common/agent-loop-policy.md and common/verification-loop-rules.md when measurable iteration or bounded retry is required.
  8. Read relevant project overlays when present: stack, architecture, styling, verification, state, data-fetching, build, and workspace profiles.
  9. Read the Design Implementation Spec, Design Direction Contract, and supplied visual references.
  10. Read affected source files, styles, components, routes, and configs.
  11. Read references/implementation-rules.md when present.

Do not read generated dist/** during normal runtime.

When shared UI or compatible component wrappers are in scope, read common/frontend-design-system-rules.md and common/component-substitution-rules.md. Apply purpose-specific naming from common/frontend-implementation-boundaries.md to changed project-owned code.

When React/Next work changes loading, imports, subscriptions or expensive rendering, read the relevant section of common/react-performance-rules.md.

Tool Contract

  • Use Project Context MCP when available; otherwise read project/** and source files directly.
  • Use Design Spec MCP when available; otherwise use the supplied spec text or user-approved local artifact.
  • Use Visual Reference MCP when available; otherwise use attached or local image files supplied by the user.
  • Use context7 for React, Next.js, Redux, TanStack, Axios, TypeScript, or build-tool docs when implementation depends on current behavior.
  • Activate openai_platform_docs only when current OpenAI API or ChatGPT Apps SDK behavior affects the implementation.
  • Use mdn for current HTML, CSS, Web API, accessibility, and compatibility facts.
  • Use rendered visual QA only when screenshot comparison, viewport evidence, overflow checks, or visible state verification are in scope.
  • Use Visual Diff MCP when available during final visual comparison.
  • Route live design reads through design-screenshot-spec and reuse its evidence.
  • If a named MCP is unavailable, report the missing capability before using a lower-confidence fallback.

Workflow

  1. Confirm the task is within the supported target stack.
  2. Detect React/Next.js routing, CSS Modules ownership, Redux ownership, TanStack usage, Axios API adapter boundaries, and verification commands.
  3. Map the spec to existing project components, layout primitives, tokens, assets, styles, and UX patterns.
  4. Check component coverage and the spec's product decision register before editing. Use skills/design-screenshot-spec/references/product-behavior-review.md through the owning skill for unresolved design or behavior. Ask the user to resolve every needed product choice; implement only confirmed independent scope. Do not invent transitions, form opening/dismissal, persistence, hidden states, responsive choices, or animation defaults to complete the page.
  5. Use frontend-architecture-planner before editing when route, state, data, form, build, workspace, or shared component ownership is unclear or material.
  6. Use loop-workflow-planner before editing when the task requires bounded retry, independent review, loop memory, or measurable iteration beyond one implementation pass.
  7. Plan component decomposition before editing.
  8. Map structure as information before adding numbers, dividers, badges, labels, grids, metrics, or decorative sections.
  9. Identify conditional UX gates: forms, navigation, data visualization, icon quality, and mobile responsiveness.
  10. Identify CSS Modules ownership and selector risks before styling.
  11. Implement the smallest scoped change that satisfies the spec, design direction, architecture handoff, and decomposition plan.
  12. Keep UI state local unless existing React, Redux, or TanStack ownership requires otherwise.
  13. Keep styles in CSS Modules by default.
  14. Reuse existing Axios clients and API adapters for API-facing changes.
  15. Do not add dependencies, global tokens, UI libraries, or new styling systems without explicit approval.
  16. Preserve accessibility, semantic structure, keyboard/focus behavior, text wrapping, responsive behavior, and stable layout dimensions.
  17. Use frontend-linter-manager after code-changing work when a lint command exists.
  18. Run relevant project verification commands from project/verification-profile.md.
  19. Run frontend-visual-qa only when the change needs rendered screenshot comparison, viewport evidence, overflow verification, or visible state verification.
  20. Use frontend-quality-reviewer before final reporting when the user asks for review, the implementation has significant surface area, a loop contract requires independent review, or the plan requires a quality pass.

Output Contract

Final response: return only facts that affect the user's understanding, confidence, or next action. Omit empty fields and workflow narration.

Report:

  • implemented spec scope;
  • affected React or Next.js surface;
  • decomposition applied or why the touched component remained intentionally small;
  • CSS Modules, Redux, TanStack, Axios, or project primitives reused;
  • form, navigation, data, icon, and responsive gates considered or skipped as not applicable;
  • loop contract followed or why no loop was needed;
  • deviations from the design spec;
  • lint and verification commands run;
  • rendered visual QA result, skipped reason, or blocker.

Domain Terms

When affected product terms have a glossary, load that domain under common/domain-glossary-rules.md. Preserve confirmed meanings across criteria, UI labels, and code identifiers. Do not silently resolve material ambiguity.

Functional Acceptance

For behavior changes, verify the user outcome under common/verification-loop-rules.md. A named browser-dependent functional criterion permits a scoped browser check under common/rendered-visual-verification-policy.md even without a visual change. Keep visual QA routed separately; report unverified runtime behavior honestly.

Validation Gates

  • Implementation must follow inspected project conventions.
  • The target must be within React, Next.js, CSS Modules, Redux, TanStack, or Axios scope.
  • CSS Modules changes must have clear ownership and stable specificity.
  • Structural devices must encode meaning instead of decoration.
  • Form, navigation, data, icon, and mobile risks must be handled when present.
  • Bounded iteration must have measurable criteria, attempt limit, and stop conditions.
  • Source evidence and confirmed user decisions must cover the implemented slice; unresolved dependent behavior is blocked, not silently filled from conventions.
  • No new package, styling system, global token, architecture layer, or UI library may appear without approval.
  • Code-changing implementation must run lint when an existing lint command is available.

Trigger Evals

Should trigger:

  • "Implement this Design Implementation Spec in the current React page."
  • "Build this approved design direction with the existing CSS Modules."

Should not trigger:

  • "Write a visual direction before coding."
  • "Compare the rendered page with the screenshot only."

Reference Map

  • common/react-performance-rules.md
  • common/target-stack-policy.md
  • skills/design-screenshot-spec/references/product-behavior-review.md
  • common/approved-patterns.md
  • common/anti-patterns.md
  • common/css-modules-specificity-rules.md
  • common/form-feedback-rules.md
  • common/navigation-ux-rules.md
  • common/data-visualization-rules.md
  • common/icon-quality-rules.md
  • common/mobile-responsive-rules.md
  • common/agent-loop-policy.md
  • common/verification-loop-rules.md
  • skills/loop-workflow-planner/SKILL.md
  • skills/frontend-architecture-planner/SKILL.md
  • skills/frontend-linter-manager/SKILL.md
  • skills/frontend-quality-reviewer/SKILL.md
  • skills/frontend-visual-qa/SKILL.md

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.