Executing Plans

Execute plans in batches with review checkpoints.

Announce: "I'm using the executing-plans skill to implement this plan."

Prerequisites

• Plan exists in docs/plans/YYYY-MM-DD-<topic>.md
• On a feature branch

The Flow

Verify Branch → EC Search → Load Plan → Execute Batches → Finish

Step 1: Verify Branch

git branch --show-current

Must be on a feature branch, not main.

Step 2: Load Context

Get project config and relevant context:

ec_search:
  query: project config
  type: config

ec_search:
  query: [feature area]
  type: pattern

ec_search:
  query: [feature area]
  type: learning

Note any gotchas or patterns that apply to this implementation.

Step 3: Load and Review Plan

1. Read the plan file
2. Choose ONE progress tracking approach (don't mix):
   • Tasks (preferred): If TaskCreate/TaskUpdate tools are available, use Tasks. Creates persistent, shareable progress.
   • TodoWrite (fallback): If Tasks aren't available, use TodoWrite for the session.
3. If concerns about the plan, raise them before starting

Important: Pick one approach and stick with it for the entire execution. Don't mix Tasks and TodoWrite.

Option A: Using Tasks (Preferred)

If TaskCreate/TaskUpdate/TaskList tools are available:

TaskCreate: "Batch 1: [first 3 tasks summary]"
TaskCreate: "Batch 2: [next 3 tasks summary]" → addBlockedBy: [batch 1 id]
TaskCreate: "Batch 3: [final tasks summary]" → addBlockedBy: [batch 2 id]

Benefits:

• Progress survives context switches and session restarts
• Subagents can share the same task list with CLAUDE_CODE_TASK_LIST_ID
• Dependencies prevent out-of-order execution

Option B: Using TodoWrite (Fallback)

If Tasks aren't available, create a TodoWrite with all batches and update as you go.

Step 4: Execute in Batches

Default batch size: 3 tasks

Before each batch, ask:

{
  "questions": [{
    "question": "How should I execute Batch N (tasks X-Y)?",
    "header": "Execution",
    "options": [
      { "label": "Main thread", "description": "Execute here with full visibility" },
      { "label": "Subagent", "description": "Fresh context, returns summary" }
    ],
    "multiSelect": false
  }]
}

Main Thread Execution

For each task:

1. Mark batch as in_progress (TaskUpdate if using Tasks, otherwise TodoWrite)
2. Follow each step exactly as written (tasks marked @tdd use that skill)
3. Run verifications as specified (@verifying)
4. Mark as completed when batch passes verification

Subagent Execution

Dispatch with this prompt:

Execute tasks X-Y from this plan:

[Paste relevant task sections from plan]

Requirements:
- Follow @tdd for each task (RED → GREEN → REFACTOR)
- Use @verifying before claiming any step complete
- Commit after each task
- Stop and report if any verification fails

EC Context:
- Test command: {test_command}
- [Relevant patterns/learnings from EC]

Return:
- Summary of what was implemented
- Files created/modified
- Any issues or blockers encountered

After subagent returns:

1. Review the summary
2. Update progress tracking (TaskUpdate if using Tasks, otherwise TodoWrite)
3. If issues reported, switch to main thread for fixes

Tip: If using Tasks, subagents can share the same task list by setting CLAUDE_CODE_TASK_LIST_ID - updates broadcast across sessions.

After each batch:

"Completed tasks N-M. [Brief summary]. Ready for feedback."

Wait for feedback before continuing.

Step 5: Request Code Review @requesting-review

After all tasks complete, use @requesting-review to:

1. Verify tests pass
2. Prepare context for reviewer
3. Dispatch code-reviewer agent
4. Handle feedback with @receiving-review

Address any Critical or Important issues before proceeding.

Step 6: Store Patterns

If the plan noted patterns to store:

ec_add:
  type: pattern
  area: [component]
  content: [Pattern description]
  rationale: Established during [feature] implementation

Step 7: Finish

When all tasks complete and review passes:

"Implementation complete. Ready to finish the branch?"

If yes → Use @finishing-branch

When to Stop

Stop and ask when:

• Blocker mid-batch (missing dependency, test fails)
• Plan has gaps
• Instruction is unclear
• Verification fails repeatedly

Don't guess - ask for clarification.