Frontend Feature logo

Frontend Feature

Organization
vstorm-co
frontend-feature

Build or change a page, view or data-driven feature in the Next.js frontend — a route under the dashboard, wiring UI to a backend endpoint, a Zustand store, a permission that must hide a control, or localized copy. Use for any work in frontend/src, including "the page renders but shows an empty state" and "this button should not be visible to a Viewer".

Overview

Publishervstorm-co
Repositoryagenticos
Skill namefrontend-feature
Stars
137
Forks
27
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by vstorm-co on GitHub. Read the source before you install it.

Installation

Install the Frontend Feature AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/vstorm-co/agenticos/tree/main/.claude/skills/frontend-feature
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/vstorm-co/agenticos.git /tmp/agenticos
mkdir -p .claude/skills
cp -r /tmp/agenticos/.claude/skills/frontend-feature .claude/skills/frontend-feature
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Frontend Feature in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Frontend Feature on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Frontend Feature is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Frontend — Next.js and React

Use the versions declared in frontend/package.json and the lockfile.

.claude/rules/frontend.md has the conventions. docs/architecture.md has the request path. This is the working layout and the traps.

Layout

Path
src/app/[locale]/(dashboard)/…The product: agents, chat, rag (knowledge bases + search; kb only redirects there), skills, vault, mcp-servers, runs, orgs, settings, admin, invitations, profile
src/app/[locale]/(auth)/…, auth/, onboarding/, legal/, shared/Everything else
src/app/api/…Route handlers that proxy to the backend (same-origin auth cookies)
src/lib/Typed API clients on api-client.ts, server-api.ts, query-keys.ts
src/hooks/One per resource: use-agents, use-skills, use-permissions, …
src/stores/Zustand, one per concern, re-exported from index.ts
src/components/<domain>/UI by domain; primitives in components/ui/, empty/error in components/states/

Routes are locale-prefixed. There is no (marketing) group.

Steps

  1. Page — src/app/[locale]/(dashboard)/<feature>/page.tsx. Server Component by default; "use client" only for state, effects or handlers.
  2. Client — a typed module in src/lib/<feature>-api.ts built on api-client.ts. Never scatter fetch in components. Server-side reads go through server-api.ts.
  3. Query keys — register in src/lib/query-keys.ts so invalidation stays consistent.
  4. Hook — src/hooks/use-<feature>.ts wrapping the client, exported from hooks/index.ts. Components consume hooks, not clients.
  5. Store — only for UI/ephemeral state. Server data belongs in the query layer.
  6. Components — under src/components/<domain>/, composed from components/ui/. Under ~100 lines; extract when they grow.
  7. i18n — every user-facing string through next-intl. Never hardcode copy.
  8. Verify:
    bash
    cd frontend && bun run type-check && bun run lint && bun run test

Permissions hide controls, and that needs a test

use-permissions.ts reads the effective permission set for the active organization. A control the caller may not use should not be rendered — not rendered-and-then-403.

Write the integration test. *.integration.test.tsx runs Testing Library against a mocked API and is the cheap, precise place to assert that a permission actually hides a button and that a form submits what it claims to. See components/agents/capability-settings.integration.test.tsx, components/teams/members-table.integration.test.tsx, components/mcp/mcp-server-list.integration.test.tsx.

This is not E2E work. Reserve Playwright for journeys crossing the whole stack — see the e2e-tests skill.

The empty-state trap

Every dashboard page fans out to several queries and renders its empty state when a query fails. "No skills yet" and "the skills request answered 502" are the same pixels. So:

  • When a page looks empty, check the network tab before the component.
  • A test that asserts a heading and a button passes against a backend that was never started. Assert on data.

Tests

bash
bun run test          # vitest watch
bun run test:run      # once
bun run test:coverage
bun run test:e2e      # playwright — see the e2e-tests skill

Unit tests sit beside the source (agent-spec.test.ts, auth-store.test.ts); integration tests are *.integration.test.tsx.

Generated files

src/lib/brand-glyphs.generated.ts is written by bun run gen:brand-icons, which fetches every brand mark the console draws — services and connectors from Simple Icons and Font Awesome, model providers from @lobehub/icons-static-svg — and bakes them in as raw SVG path data. It is the only source of a brand mark: BrandIcon, brandMark(name) and ProviderIcon read it, and nothing else. Adding a mark is a row in BRANDS or PROVIDERS in scripts/gen-brand-icons.ts plus a re-run — not an icon package, and not a hand-authored d. This replaced react-icons and @lobehub/icons in #156: 199 MB installed for 89 marks whose artwork the generator fetches from the same upstream sets, byte for byte.

src/lib/mcp-logos.generated.ts is written by bun run gen:mcp-logos, which fetches each catalog server's favicon and bakes it in as a base64 data URI so the demo MCP badge renders in a self-contained HTML export with no network. Do not hand-edit it — regenerate after changing mcp_servers.json.

Neither is hand-edited, and make lint will not catch it if you do — the guard is that both files say so at the top and both regenerate deterministically, so a hand edit is lost on the next run.

That is separate from the backend icon contract (app/core/catalog/icons/<name>.svg, served as a currentColor silhouette) — see the project-docs and mcp-connections skills.

Frequently asked questions

What does the Frontend Feature AI skill do?

Build or change a page, view or data-driven feature in the Next.js frontend — a route under the dashboard, wiring UI to a backend endpoint, a Zustand store, a permission that must hide a control, or localized copy. Use for any work in frontend/src, including "the page renders but shows an empty state" and "this button should not be visible to a Viewer".

Why use Frontend Feature on TypingMind?

Because you install it once and use it with any model. Frontend Feature is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Frontend Feature in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vstorm-co/agenticos/tree/main/.claude/skills/frontend-feature. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Frontend Feature?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Frontend Feature?

As many as you like. As long as a model supports skills, you can use Frontend Feature with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Frontend Feature AI skill free?

Yes. It is published on GitHub by vstorm-co under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇