agentleFS
openvettaopenvetta/open-vetta/.agents/skills/compose-state-and-effects/SKILL.md

compose-state-and-effects

Skill299 starsChanged 18 days agoApache-2.0View on GitHub

Use when writing or reviewing Jetpack Compose state ownership, remember state, state hoisting, screen state holders, LaunchedEffect, DisposableEffect, SideEffect, Flow collection, navigation, snackbar, analytics, or focus requests.

2 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Place each state value at its lowest necessary owner: local UI state, hoisted state, a plain UI state holder, or a screen state holder.

    line 25
  2. Very specific:

    Keep this content callable with sample state and callbacks.

    line 32
  3. Distinctive:

    Before adding it, confirm the module declares its tooling.

    line 34
  4. Distinctive:

    Before responding to a screen-ownership review, verify all three screen seams in step 4 when visible code needs them.

    line 47

≈ “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

Compose state and effects

Core principle

Give every piece of UI state one lowest responsible owner, then run imperative work through the effect whose lifecycle follows that owner. Composition renders; state and effects make rendering change safely.

A screen-ownership review is incomplete until it names the plain, previewable content composable that receives immutable state and event callbacks separately from app wiring and composition-owned runtime objects.

Procedure

  1. Establish the requested scope and visible behavioral requirements. Treat an ownership change as a finding only when code or task evidence shows a lifecycle, testability, business, or coordination need.
  2. Inventory mutable UI state, app state, event streams, app dependencies, and imperative work in the affected screen or component.
  3. Place each state value at its lowest necessary owner: local UI state, hoisted state, a plain UI state holder, or a screen state holder.
  4. For a screen boundary, keep durable data and intents in the wiring owner, Compose runtime objects in composition or a plain UI state holder, and rendering in a previewable content composable that takes immutable state and callbacks. Read State hoisting for the implementation shape; naming only state and intents is not that boundary. Keep this content callable with sample state and callbacks; an @Preview wrapper is optional and is not required for the state-driven boundary. Before adding it, confirm the module declares its tooling; do not assume Compose runtime or UI dependencies provide it. Respect the task's file scope: do not add tooling or change build files unless authorized. If Preview tooling is needed but unavailable within scope, omit the wrapper and report the setup gap.
  5. Choose an effect API whose lifecycle matches the work, and key it by the semantic input that should restart or dispose it.
  6. Load the focused reference for every material concern below. Do not use a reference merely because its topic is adjacent.
  7. Route frame-rate reads, cross-phase back-writing, and @ReadOnlyComposable contracts to Compose performance. For that contract, distinguish a read of snapshot State.value from an operation that changes the composer; the state read alone is not a violation.
  8. Before responding to a screen-ownership review, verify all three screen seams in step 4 when visible code needs them.
  9. Finish when every state value has one owner, every effect has a justified lifecycle and key, and the UI can be previewed and tested without app dependencies. For review-only work, report no change when no evidence-backed issue remains; do not invent product requirements.

Topic router

SignalRead
Bare local var, remember { mutableStateOf(...) }, state lists/maps, or reset stateLocal state
State shared by siblings, UI state holders, ViewModel/component wiring, or previewable screen boundariesState hoisting
LaunchedEffect, DisposableEffect, SideEffect, snapshotFlow, rememberCoroutineScope, rememberUpdatedState, produceState, imperative requestFocus, callbacks, event Flow collection, snackbar, navigation, or analyticsSide effects
Focus ownership and keyboard/TV/D-pad behaviorCompose focus navigation
Tests or previews for the resulting UI contractCompose UI testing patterns

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.