agentleFS
Joyboy-dyJoyboy-dy/felicio-ai-skills/skills/senior-frontend-engineering/SKILL.md

senior-frontend-engineering

Skill0 starsChanged 3 months agoMITView on GitHub

Use when implementing or refactoring substantial frontend features with production-quality component ownership, state/data flow, API contracts, accessibility, testing, performance, error handling, and maintainability across modern frontend frameworks.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 6 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

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

    Use a framework-specific skill alongside this one when available.

    line 14
  2. Distinctive:

    Keep state as local as possible and elevate it only when multiple owners need the same source of truth.

    line 20
  3. Distinctive:

    Keep server/API validation and authorization authoritative.

    line 23
  4. Very specific:

    Do not add a dependency when platform/framework/existing utilities solve the task cleanly.

    line 24
  5. Distinctive:

    Avoid speculative abstractions and premature global state.

    line 25
  6. Very specific:

    Use existing tokens/primitives/variants.

    line 43
  7. Very specific:

    Prefer derived values over synchronized duplicate state.

    line 47
  8. Distinctive:

    Keep layout stable where practical.

    line 51
  9. Distinctive:

    Use semantic elements, labels/names, visible focus, keyboard operability, appropriate announcements, and correct disabled/read-only …

    line 55
  10. Very specific:

    Run available formatter/lint/typecheck/tests/build and exercise representative runtime paths when possible.

    line 63
  11. Distinctive:

    Use the health scanner for fast repository triage:

    line 65

≈ “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

Senior Frontend Engineering

Implement frontend work as maintainable product code, adapting to the repository's actual framework and conventions.

Use when

Activate for non-trivial frontend implementation/refactoring, component boundaries, state ownership, async data flows, API integration, design-system integration, dependency choices, error/loading states, frontend test strategy, or maintainability/performance review.

Use a framework-specific skill alongside this one when available. nextjs-architecture-pattern owns Next.js-specific routing/rendering details. frontend-aesthetics owns visual art direction. form-ux-validation owns detailed form behavior.

Operating rules

  • Inspect before implementing: framework, package manager, source layout, components, styling, state/data libraries, API client, tests, lint/type/build scripts.
  • Preserve existing sound conventions instead of importing a preferred personal stack.
  • Keep state as local as possible and elevate it only when multiple owners need the same source of truth.
  • Reuse stable primitives and tokens before creating near-duplicates.
  • Model async states explicitly: idle/loading/success/empty/error/stale as relevant.
  • Keep server/API validation and authorization authoritative; frontend checks improve UX, not trust.
  • Do not add a dependency when platform/framework/existing utilities solve the task cleanly.
  • Avoid speculative abstractions and premature global state.

Workflow

1. Establish ownership

Identify which component/module owns the data, mutation, UI state, and side effects. Read references/architecture-and-ownership.md for ambiguous cases.

2. Trace contracts

Inspect the actual API/types/schema and error behavior. Determine input/output/nullability, pagination, permissions, loading/retry/cancellation behavior, and cache invalidation. Do not infer a contract from UI labels.

3. Design component boundaries

Split where responsibility, reuse, testing, rendering boundary, or state ownership warrants it. Avoid one-file mega-components, but do not atomize trivial markup into wrappers with no independent meaning.

4. Integrate the design system

Use existing tokens/primitives/variants. Add reusable variants at the appropriate primitive layer when several consumers need them. Read references/design-system.md for integration rules.

5. Handle state and effects deliberately

Prefer derived values over synchronized duplicate state. Effects are for synchronization with external systems, not for ordinary value derivation. Clean up subscriptions/timers/requests when the framework requires it. Prevent stale async writes when race conditions are possible.

6. Make errors and loading useful

Keep layout stable where practical. Preserve prior useful data during refresh when the product expects it. Give actionable recovery for recoverable errors. Do not swallow exceptions or turn all errors into an undifferentiated toast.

7. Accessibility and keyboard behavior

Use semantic elements, labels/names, visible focus, keyboard operability, appropriate announcements, and correct disabled/read-only semantics. Component-library usage does not automatically prove accessibility.

8. Test risk, not implementation trivia

Prioritize tests around contracts, user-visible state transitions, regressions, permissions/edge cases, and reusable logic. Avoid snapshot-only confidence for behavior-heavy features.

9. Validate end-to-end

Run available formatter/lint/typecheck/tests/build and exercise representative runtime paths when possible. Read references/guardrails-and-workflow.md before broad or high-risk refactors.

Use the health scanner for fast repository triage:

python scripts/frontend_health.py <project-root> --json

Common failure modes

  • Replacing an existing library because another is more familiar.
  • Copying API types by hand into several UI files.
  • Globalizing modal/filter/form state with one consumer.
  • Effects that continuously synchronize state that could be derived.
  • Duplicate Button/Card/Dialog variants scattered across feature folders.
  • Happy-path-only UI with no empty/error/pending handling.
  • Optimistic UI that treats local state as authoritative success.
  • Huge refactor bundled into an unrelated feature.
  • Accessibility checked only by visual appearance.
  • “Performance optimization” without measuring or identifying an actual hot path.

Output

State detected stack/conventions, ownership and contract decisions, implementation changes, user-visible states covered, dependencies added/avoided, validations actually run, and remaining runtime uncertainty.

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.