Iteration Tracking

This skill provides iteration state management, progress tracking, and checkpoint creation for development cycles.

What This Skill Provides

1. Iteration State Management

• Initialize new iteration cycles
• Track iteration status and progress
• Manage iteration metadata

2. Checkpoint System

• Create mid-iteration checkpoints
• Capture git state at checkpoints
• Track progress between checkpoints

3. Iteration Reports

• Progress summaries
• Change tracking
• Time and effort metrics

Instructions

Starting an Iteration

When user wants to start a new iteration:

1. Create iteration directory structure: !{bash mkdir -p .multiagent/iterations/$ITERATION_NAME}

2. Initialize state file with JSON:

   {
     "name": "$ITERATION_NAME",
     "status": "in_progress",
     "started_at": "timestamp",
     "checkpoints": []
   }
   

3. Create symlink to current iteration: !{bash ln -sf $ITERATION_NAME .multiagent/iterations/current}

Creating Checkpoints

When user wants to create a checkpoint:

1. Load current iteration
2. Create checkpoint directory
3. Capture git status and diff
4. Update state with checkpoint metadata

Ending an Iteration

When user completes an iteration:

1. Create final snapshot
2. Update state to "completed"
3. Remove current symlink
4. Generate summary report

State File Structure

{
  "name": "feature-auth",
  "status": "in_progress",
  "started_at": "2025-10-24T10:00:00Z",
  "checkpoints": [
    {
      "id": "1234567890",
      "timestamp": "2025-10-24T11:00:00Z",
      "message": "Initial implementation"
    }
  ],
  "completed_at": null
}

Success Criteria

• ✅ Iteration state is persisted and trackable
• ✅ Checkpoints capture meaningful progress
• ✅ Reports show clear progress metrics
• ✅ State transitions are clean and consistent

Plugin: 04-iterate Skill Type: State Management + Tracking Auto-invocation: Yes (via description matching)