Agent Skills › lobehub/lobehub › compose-atoms

compose-atoms

GitHub

指导将重型领域模块按能力拆分为原子组件,支持下沉状态与按需挂载,避免简单文件分割。

.agents/skills/compose-atoms/SKILL.md lobehub/lobehub

Trigger Scenarios

需要拆分重型领域模块为独立能力单元 多宿主环境下需优化模块加载与状态管理

Install

npx skills add lobehub/lobehub --skill compose-atoms -g -y
More Options

Non-standard path

npx skills add https://github.com/lobehub/lobehub/tree/canary/.agents/skills/compose-atoms -g -y

Use without installing

npx skills use lobehub/lobehub@compose-atoms

指定 Agent (Claude Code)

npx skills add lobehub/lobehub --skill compose-atoms -a claude-code -g -y

安装 repo 全部 skill

npx skills add lobehub/lobehub --all -g -y

预览 repo 内 skill

npx skills add lobehub/lobehub --list

SKILL.md

Frontmatter
{
    "name": "compose-atoms",
    "description": "Use for splitting heavy domains into capabilities (原子组件) across page, portal, share or micro-app hosts, sinking state (状态下沉) and replacing mode\/readOnly flags. Simple component splits: react.",
    "user-invocable": false
}

Compose Atoms

A heavy domain is a kit, not a viewer with modes. Each host is an import list. An atom owns the data, actions, and dependencies of one capability. The assembler only chooses what to mount.

This is a module-graph split. Hiding UI does not drop a module. A host that never imports a file is the only host that does not ship it.

Do not use this skill to slice one component into smaller files. That is react.

When

The folder is a product surface (Conversation, Acceptance, Document, Verify, Task, …) and at least one of these is true:

  • One entry owns read + write + workflow + host integration.
  • A second host already exists or is the next change (page, portal, share, mobile, popup, micro-app).
  • The next feature is another boolean / mode / variant on the fat tree.
  • Reusing a header, list, or card would import trays, stores, or chat.

If the file is large but has one capability and one host, stop. Use react.

Grain

An atom is the smallest unit a host is allowed not to mount.

Ask, for every chunk: would any host ship the rest and skip this? If yes, it is an atom. If no, it stays inside its parent.

That grain is coarser than a visual section and finer than the whole page.

Too coarse Right grain Too fine
One Viewer with readOnly Identity, goal, list, decision, … Every badge, row, and icon as a public export
Chat as a single import List, composer, intervention, … Every message sub-row the page then rewires
Document as editor + every panel Canvas, header, comments as omitted Internal toolbar buttons

A workflow (focus review, thread, publish) is one atom to hosts that skip the whole mode. Inside it, keep splitting only if a host would mount a piece of that workflow alone.

Sink State

The split is worthless if the assembler still holds the domain state. Imports follow the hook. A useStore / useX / handleAccept left on the page keeps that module (and everything it imports) on every host that mounts the page.

Sink everything the atom needs into the atom:

Sinks into the atom Stays on the assembler
Resource fetch The record id / scope identity
Derived view model (counts, labels, scope) Which atoms are mounted (the import list)
Transient UI (filter, collapse, pending) Host layout state (focused vs overview, rail)
Mutation handlers and their stores/modals Host seams (provider value the runtime owns)
// wrong — visual split, state still on the page
const Page = () => {
  const { data, mutate } = useX(id);
  const [filter, setFilter] = useState('all');
  const accept = () => void mutate(...);
  return (
    <>
      <Identity data={data} />
      <List data={data} filter={filter} onFilter={setFilter} />
      <DecisionBar onAccept={accept} />
    </>
  );
};

// right — each atom reads and acts; page only assembles
const Page = () => (
  <Scope id={id}>
    <Identity />
    <List />
    <Decision />
  </Scope>
);

Do not lift "to fetch once" or "so siblings share data". Two atoms calling the same SWR hook share the cache by key. Sibling writes go through that cache (or a store slice the write atom imports), not through page useState.

Lift only when the state is the assembler's job: which workflow is on screen, whether a rail is open. If a value is used to render or mutate one capability, it belongs in that capability.

Kit

Piece Owns Must not own
Domain primitive Types, predicates, formatters React, stores, services
Data access One resource, one file Sibling list / infinite / document hooks
Read atom Present + local UI state + its own fetch Mutations, trays, owner services, host stores
Write atom One mutation and its UI Page layout, unrelated writes
Workflow atom A mode assembled from other atoms Becoming the only entry other hosts can import
Domain slot ReactNode hole for another same-domain atom A callback whose implementation lives in this file
Host seam Optional host-provided UI, null by default A fallback that imports the host graph
Assembler Layout + the import list Domain actions; domain fetch except local chrome

A visual block that both reads and writes is a read atom + a write atom, joined by a domain slot.

Two seam types — do not mix them:

  • Domain slot — same feature, optional capability (editSlot, toolbar). The host imports the write atom and passes it in.
  • Host seam — only some runtimes can resolve it (chat store, topic drawer, Electron). Context default is null. The micro-app / share page never imports the provider value. See split-micro-app for the runtime cut.

Ownership

State and imports live in the atom that needs them.

  • Read atoms call the isolated resource hook. Filter / collapse / expand stay in the atom.
  • Write atoms call the same hook, then mutate. They import their own modals and services. They return null when the record is not writable.
  • Workflow atoms compose other atoms. They still do not become the light host's entry.
  • Assemblers do not fetch in order to push props down. Chrome unique to that host (a title in a shell header) may read the hook.
  • Scope context carries identity (id, embedded). It does not carry actions, stores, or modal openers.

Several atoms calling the same SWR hook is expected. SWR dedupes on key. Do not lift data to the assembler "to fetch once".

The shared read hook is its own file. A light host imports that file, never a hooks.ts / store barrel that also exports list, document, or send-message.

// any host:
import { useX } from '@/features/Domain/useX';

// never from a light host:
import { useX } from '@/features/Domain/hooks';
import { DomainViewer } from '@/features/Domain';

A flag that only changes behavior of code already in the atom is fine. A flag whose purpose is to skip importing another module is not — that module is a slot or a sibling the assembler omits.

dynamic(() => import(...)) only helps when no static import of that module remains on the light graph.

Assembler

A host is its import list.

// light host — deep imports only
<Scope id={id}>
  <ReadA />
  <ReadB />
</Scope>

// full host — same reads, write atoms in slots, workflows as siblings
<ReadA extra={<WriteA />} />
<ReadB toolbar={<WriteB />} />
<Workflow />
<WriteC />

Route / portal files compose host skeleton (providers, host seams). Do not add a wrapper page whose only job is to sit between the route and the assembler.

The feature barrel may export the full assembler for in-app use. Light hosts must not import that barrel.

Procedure

  1. List hosts (page, portal, share, mobile, popup, micro-app — include the next one).
  2. List capabilities as verbs (view identity, edit field, decide, open host thing). Map host → verbs.
  3. Set grain: each verb a host can skip is an atom. Group the rest under a parent.
  4. Extract primitives and the isolated read hook first.
  5. Extract read atoms. Optional same-domain capabilities become ReactNode slots, not callbacks that close over write modules.
  6. Extract each write / workflow atom into its own file. Heavy deps stay in that file.
  7. Sink state. Move fetch, view-model, transient UI, and handlers out of the assembler. If the page still calls a domain hook to feed props, the split is not done.
  8. Rewrite every host as an import list plus layout. Delete readOnly / isPublic / mode / variant on the fat tree.
  9. Point light hosts at atoms by deep path.
  10. Prove the cut. Grep the light entry's import tree, or trace a module the light host must not ship. A missing button is not proof.
# example: light host must not import a write atom
WORKBENCH_TRACE_MODULE=features/Acceptance/Viewer/Review/AcceptanceDecision bun run build:rr

If the light assembler appears on the importer chain, a static import still exists — usually a barrel, a leftover flag, or actions left on the page.

Example

src/features/Acceptance/Viewer/ is one application of this kit, not the template to copy file-for-file.

Host Mounts
In-app Acceptance/index.tsx Identity, goal, inventory, decision, focus workflow, ledger + write slots
Workbench public detail Identity, goal, inventory, discussion with sign-in guidance
Portal Full assembler + OriginConversationProvider

Read AcceptanceGoal (read + editSlot) and AcceptanceGoalEdit (write) for the slot cut. Read originConversation.tsx for a host seam. Read workbench AcceptanceDetail.tsx for a light assembler.

Related skills

  • react: single-surface boundaries, styling, memoization.
  • split-micro-app: worker / SSR / gateway / stubs. This skill is how the shared UI is cut afterwards.
  • spa-routes: route files stay thin assemblers.
  • data-fetching-architecture: how the isolated hook fetches, not where a host may import it from.
  • zustand: narrow selectors inside an atom; do not put the whole store on scope context.

Version History

  • 69012b8 Current 2026-09-28 20:13
  • e0d4b61 2026-09-23 09:51
  • a06b4e2 2026-08-29 05:59
  • 29fe043 2026-08-20 18:33

Same Skill Collection

.agents/skills/add-model-provider/SKILL.md
.agents/skills/add-provider-doc/SKILL.md
.agents/skills/add-setting-env/SKILL.md
.agents/skills/agent-runtime-hooks/SKILL.md
.agents/skills/agent-signal/SKILL.md
.agents/skills/agent-testing-bot/SKILL.md
.agents/skills/agent-tracing/SKILL.md
.agents/skills/agent-work/SKILL.md
.agents/skills/alint-ruleset/SKILL.md
.agents/skills/builtin-tool/SKILL.md
.agents/skills/chat-sdk/SKILL.md
.agents/skills/cleanup-git-worktrees/SKILL.md
.agents/skills/cli/SKILL.md
.agents/skills/data-fetching-architecture/SKILL.md
.agents/skills/db-migrations/SKILL.md
.agents/skills/debug-frontend-with-browser/SKILL.md
.agents/skills/debug-package/SKILL.md
.agents/skills/deep-review/SKILL.md
.agents/skills/design-prototype/SKILL.md
.agents/skills/desktop-security-update/SKILL.md
.agents/skills/desktop/SKILL.md
.agents/skills/docs-changelog/SKILL.md
.agents/skills/drizzle/SKILL.md
.agents/skills/full-text-search/SKILL.md
.agents/skills/heterogeneous-agent/SKILL.md
.agents/skills/hotkey/SKILL.md
.agents/skills/i18n/SKILL.md
.agents/skills/linear/SKILL.md
.agents/skills/llm-generation/SKILL.md
.agents/skills/modal/SKILL.md
.agents/skills/model-bank-metadata/SKILL.md
.agents/skills/product-design/SKILL.md
.agents/skills/project-overview/SKILL.md
.agents/skills/react/SKILL.md
.agents/skills/response-compliance/SKILL.md
.agents/skills/run-eval-harbor/SKILL.md
.agents/skills/skills-audit/SKILL.md
.agents/skills/source-layout/SKILL.md
.agents/skills/spa-routes/SKILL.md
.agents/skills/split-large-file/SKILL.md
.agents/skills/split-micro-app/SKILL.md
.agents/skills/store-data-structures/SKILL.md
.agents/skills/testing-heterogeneous-agents/SKILL.md
.agents/skills/testing/SKILL.md
.agents/skills/trpc-router/SKILL.md
.agents/skills/typescript/SKILL.md
.agents/skills/upstash-workflow/SKILL.md
.agents/skills/ux-audit/SKILL.md
.agents/skills/ux/SKILL.md

Metadata

Files
0
Version
aace12f
Hash
6d4e33b6
Indexed
2026-08-20 18:33

inicio - Wiki
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-10-27 10:25
浙ICP备14020137号-1