agentleFS
levyfluxlevyflux/codex-skills/skills/flutter-flow-audit/SKILL.md

flutter-flow-audit

Skill1 starsChanged 3 months agoMITView on GitHub

Audit a Flutter user flow across routes, screens, state owners, services, SDK callbacks, async boundaries, and lifecycle cleanup. Use for read-only diagnosis of cross-screen state bugs, stale UI, duplicated subscriptions, navigation races, reconnect behavior, or business-flow regressions; do not use static inspection alone as runtime proof.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read invariants-and-transitions.md while constructing the model.

    line 10
  2. Distinctive:

    Write the expected state machine and invariants before labeling findings.

    line 18
  3. Distinctive:

    Check every transition's producer, consumer, ordering, idempotency, cancellation, error path, and ownership.

    line 19

≈ “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

Flutter Flow Audit

Build a source-backed state-and-event model for one user flow, then classify mismatches without assuming every suspicious pattern is a defect.

Read invariants-and-transitions.md while constructing the model.

Audit workflow

  1. Establish the user-visible entry point, expected outcome, supported platforms, and authoritative behavior source.
  2. Trace routing forward from the entry and backward from the affected screen.
  3. Identify state owners: widgets, controllers, providers, BLoCs, repositories, services, SDKs, storage, and platform channels.
  4. Enumerate events and async boundaries: initialization, callbacks, streams, timers, retries, navigation, app lifecycle, reconnect, and disposal.
  5. Write the expected state machine and invariants before labeling findings.
  6. Check every transition's producer, consumer, ordering, idempotency, cancellation, error path, and ownership.
  7. Classify each item as confirmed defect, specification ambiguity, change request, accepted behavior, false positive, or unresolved.
  8. Recommend the smallest fix and the test or runtime probe that would prove it. Implement only when the user asks and intended behavior is unambiguous.

High-value checks

  • multiple instances of the same logical state owner;
  • subscriptions installed more than once or disposed by the wrong owner;
  • callbacks that outlive a widget, controller, account, or route;
  • setState or notifier updates after disposal;
  • stale async results overwriting newer state;
  • navigation executed before persistence or cleanup completes;
  • loading, empty, error, disconnected, reconnecting, and retry states;
  • state derived independently in multiple screens;
  • platform-specific paths with different lifecycle behavior;
  • error handling that logs but leaves the UI in an impossible state.

Output contract

Provide a compact flow map, state-owner table, transition ledger, classified findings with source evidence, missing runtime evidence, and proposed validation. Distinguish static inference, test evidence, emulator/device evidence, and production evidence.

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.