asyncredux-undo-redo
Implement undo/redo functionality using state observers. Covers recording state history with stateObserver, creating a RecoverStateAction, implementing undo for the full state or partial state, and managing history limits.
Best use case
asyncredux-undo-redo is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Implement undo/redo functionality using state observers. Covers recording state history with stateObserver, creating a RecoverStateAction, implementing undo for the full state or partial state, and managing history limits.
Teams using asyncredux-undo-redo should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/asyncredux-undo-redo/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How asyncredux-undo-redo Compares
| Feature / Agent | asyncredux-undo-redo | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Implement undo/redo functionality using state observers. Covers recording state history with stateObserver, creating a RecoverStateAction, implementing undo for the full state or partial state, and managing history limits.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Undo and Redo in AsyncRedux
AsyncRedux simplifies undo/redo through a straightforward pattern: create a state observer to save states to a history list, and create actions to navigate through that history.
## Understanding StateObserver
The `StateObserver` abstract class tracks state modifications. Implement its `observe` method to be notified of state changes:
```dart
abstract class StateObserver<St> {
void observe(
ReduxAction<St> action,
St stateIni,
St stateEnd,
Object? error,
int dispatchCount,
);
}
```
**Parameters:**
- `action` - The dispatched action that triggered the state change
- `stateIni` - The state before the reducer applied changes
- `stateEnd` - The new state returned by the reducer
- `error` - Null if successful; contains the thrown error otherwise
- `dispatchCount` - Sequential dispatch number
**Timing:** The observer fires right after the reducer returns, before both the `after()` method and error-wrapping processes.
## Step 1: Create the UndoRedoObserver
```dart
class UndoRedoObserver implements StateObserver<AppState> {
final List<AppState> _history = [];
int _currentIndex = -1;
final int maxHistorySize;
UndoRedoObserver({this.maxHistorySize = 50});
@override
void observe(
ReduxAction<AppState> action,
AppState stateIni,
AppState stateEnd,
Object? error,
int dispatchCount,
) {
// Skip if action had an error
if (error != null) return;
// Skip if state didn't change
if (stateIni == stateEnd) return;
// Skip undo/redo actions to prevent recursive history entries
if (action is UndoAction || action is RedoAction) return;
// When navigating backwards then performing new actions,
// clear the "future" history
if (_currentIndex < _history.length - 1) {
_history.removeRange(_currentIndex + 1, _history.length);
}
// Add the new state to history
_history.add(stateEnd);
_currentIndex = _history.length - 1;
// Enforce maximum history size by removing oldest entries
while (_history.length > maxHistorySize) {
_history.removeAt(0);
_currentIndex--;
}
}
/// Returns the previous state, or null if at the beginning
AppState? getPreviousState() {
if (_currentIndex > 0) {
_currentIndex--;
return _history[_currentIndex];
}
return null;
}
/// Returns the next state, or null if at the end
AppState? getNextState() {
if (_currentIndex < _history.length - 1) {
_currentIndex++;
return _history[_currentIndex];
}
return null;
}
bool get canUndo => _currentIndex > 0;
bool get canRedo => _currentIndex < _history.length - 1;
}
```
## Step 2: Register the Observer with the Store
Pass the observer to `stateObservers` during store creation:
```dart
// Create the observer instance so actions can access it
final undoRedoObserver = UndoRedoObserver(maxHistorySize: 100);
var store = Store<AppState>(
initialState: AppState.initialState(),
stateObservers: [undoRedoObserver],
);
```
## Step 3: Create Navigation Actions
Create `UndoAction` and `RedoAction` that retrieve states from history:
```dart
class UndoAction extends ReduxAction<AppState> {
final UndoRedoObserver observer;
UndoAction(this.observer);
@override
AppState? reduce() {
return observer.getPreviousState();
}
}
class RedoAction extends ReduxAction<AppState> {
final UndoRedoObserver observer;
RedoAction(this.observer);
@override
AppState? reduce() {
return observer.getNextState();
}
}
```
**Alternative:** Access the observer through dependency injection using the environment pattern:
```dart
class UndoAction extends ReduxAction<AppState> {
@override
AppState? reduce() {
final observer = env.undoRedoObserver;
return observer.getPreviousState();
}
}
```
## Step 4: Integrate with the UI
Dispatch undo/redo actions from widgets:
```dart
class UndoRedoButtons extends StatelessWidget {
final UndoRedoObserver observer;
const UndoRedoButtons({required this.observer});
@override
Widget build(BuildContext context) {
return Row(
children: [
IconButton(
icon: Icon(Icons.undo),
onPressed: observer.canUndo
? () => context.dispatch(UndoAction(observer))
: null,
),
IconButton(
icon: Icon(Icons.redo),
onPressed: observer.canRedo
? () => context.dispatch(RedoAction(observer))
: null,
),
],
);
}
}
```
## Partial State Undo/Redo
The same approach works to undo/redo only **part** of the state. This is useful when you want to track changes to a specific slice of state independently.
```dart
class PartialUndoRedoObserver implements StateObserver<AppState> {
final List<DocumentState> _history = [];
int _currentIndex = -1;
final int maxHistorySize;
PartialUndoRedoObserver({this.maxHistorySize = 50});
@override
void observe(
ReduxAction<AppState> action,
AppState stateIni,
AppState stateEnd,
Object? error,
int dispatchCount,
) {
if (error != null) return;
if (action is UndoDocumentAction || action is RedoDocumentAction) return;
// Only track changes to the document portion of state
if (stateIni.document == stateEnd.document) return;
if (_currentIndex < _history.length - 1) {
_history.removeRange(_currentIndex + 1, _history.length);
}
_history.add(stateEnd.document);
_currentIndex = _history.length - 1;
while (_history.length > maxHistorySize) {
_history.removeAt(0);
_currentIndex--;
}
}
DocumentState? getPreviousDocument() {
if (_currentIndex > 0) {
_currentIndex--;
return _history[_currentIndex];
}
return null;
}
DocumentState? getNextDocument() {
if (_currentIndex < _history.length - 1) {
_currentIndex++;
return _history[_currentIndex];
}
return null;
}
}
class UndoDocumentAction extends ReduxAction<AppState> {
final PartialUndoRedoObserver observer;
UndoDocumentAction(this.observer);
@override
AppState? reduce() {
final previousDoc = observer.getPreviousDocument();
if (previousDoc == null) return null;
return state.copy(document: previousDoc);
}
}
```
## Managing History Limits
Key considerations for history management:
1. **Set appropriate limits** - Balance memory usage with undo depth needs
2. **Remove oldest entries** - When exceeding the limit, remove from the beginning
3. **Clear future history** - When new actions occur after undoing, discard the redo stack
4. **Filter irrelevant actions** - Skip actions that don't change state or are navigation actions
```dart
// Example: Different limits for different use cases
final documentObserver = UndoRedoObserver(maxHistorySize: 100); // Heavy undo
final preferencesObserver = UndoRedoObserver(maxHistorySize: 10); // Light undo
```
## Complete Example
```dart
// observer.dart
class UndoRedoObserver implements StateObserver<AppState> {
final List<AppState> _history = [];
int _currentIndex = -1;
final int maxHistorySize;
UndoRedoObserver({this.maxHistorySize = 50});
@override
void observe(
ReduxAction<AppState> action,
AppState stateIni,
AppState stateEnd,
Object? error,
int dispatchCount,
) {
if (error != null) return;
if (stateIni == stateEnd) return;
if (action is UndoAction || action is RedoAction) return;
if (_currentIndex < _history.length - 1) {
_history.removeRange(_currentIndex + 1, _history.length);
}
_history.add(stateEnd);
_currentIndex = _history.length - 1;
while (_history.length > maxHistorySize) {
_history.removeAt(0);
_currentIndex--;
}
}
AppState? getPreviousState() {
if (_currentIndex > 0) {
_currentIndex--;
return _history[_currentIndex];
}
return null;
}
AppState? getNextState() {
if (_currentIndex < _history.length - 1) {
_currentIndex++;
return _history[_currentIndex];
}
return null;
}
bool get canUndo => _currentIndex > 0;
bool get canRedo => _currentIndex < _history.length - 1;
void clear() {
_history.clear();
_currentIndex = -1;
}
}
// actions.dart
class UndoAction extends ReduxAction<AppState> {
@override
AppState? reduce() => env.undoRedoObserver.getPreviousState();
}
class RedoAction extends ReduxAction<AppState> {
@override
AppState? reduce() => env.undoRedoObserver.getNextState();
}
// main.dart
void main() {
final undoRedoObserver = UndoRedoObserver(maxHistorySize: 100);
final store = Store<AppState>(
initialState: AppState.initialState(),
stateObservers: [undoRedoObserver],
environment: Environment(undoRedoObserver: undoRedoObserver),
);
runApp(
StoreProvider<AppState>(
store: store,
child: MyApp(),
),
);
}
```
## References
URLs from the documentation:
- https://asyncredux.com/flutter/miscellaneous/undo-and-redo
- https://asyncredux.com/flutter/basics/store
- https://asyncredux.com/flutter/miscellaneous/logging
- https://asyncredux.com/flutter/miscellaneous/metrics
- https://asyncredux.com/flutter/advanced-actions/redux-action
- https://asyncredux.com/flutter/testing/store-tester
- https://asyncredux.com/flutter/basics/sync-actions
- https://asyncredux.com/flutter/advanced-actions/before-and-after-the-reducer
- https://asyncredux.com/flutter/basics/async-actionsRelated Skills
asyncredux-observers
Set up observers for debugging and monitoring. Covers implementing actionObservers for dispatch logging, stateObserver for state change tracking, combining observers with globalWrapError, and using observers for analytics.
asyncredux-nonreentrant-mixin
Add the NonReentrant mixin to prevent an action from dispatching while already in progress. Covers preventing duplicate form submissions, avoiding race conditions, and protecting long-running operations.
asyncredux-async-actions
Creates AsyncRedux (Flutter) asynchronous actions for API calls, database operations, and other async work.
bgo
Automates the complete Blender build-go workflow, from building and packaging your extension/add-on to removing old versions, installing, enabling, and launching Blender for quick testing and iteration.
mcp-create-declarative-agent
Skill converted from mcp-create-declarative-agent.prompt.md
MCP Architecture Expert
Design and implement Model Context Protocol servers for standardized AI-to-data integration with resources, tools, prompts, and security best practices
mathem-shopping
Automatiserar att logga in på Mathem.se, söka och lägga till varor från en lista eller recept, hantera ersättningar enligt policy och reservera leveranstid, men lämnar varukorgen redo för manuell checkout.
math-modeling
本技能应在用户要求"数学建模"、"建模比赛"、"数模论文"、"数学建模竞赛"、"建模分析"、"建模求解"或提及数学建模相关任务时使用。适用于全国大学生数学建模竞赛(CUMCM)、美国大学生数学建模竞赛(MCM/ICM)等各类数学建模比赛。
matchms
Mass spectrometry analysis. Process mzML/MGF/MSP, spectral similarity (cosine, modified cosine), metadata harmonization, compound ID, for metabolomics and MS data processing.
managing-traefik
Manages Traefik reverse proxy for local development. Use when routing domains to local services, configuring CORS, checking service health, or debugging connectivity issues.
managing-skills
Install, find, update, and manage agent skills. Use when the user wants to add a new skill, search for skills that do something, check if skills are up to date, or update existing skills. Triggers on: install skill, add skill, get skill, find skill, search skill, update skill, check skills, list skills.
manage-agents
Create, modify, and manage Claude Code subagents with specialized expertise. Use when you need to "work with agents", "create an agent", "modify an agent", "set up a specialist", "I need an agent for [task]", or "agent to handle [domain]". Covers agent file format, YAML frontmatter, system prompts, tool restrictions, MCP integration, model selection, and testing.