executing-plans

Executing Plans

Overview

Load plan, review critically, execute tasks in batches, report for review between batches.

Core principle: Batch execution with checkpoints for architect review.

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

Question Style (Check First)

Before starting, check if a question style was set this session (e.g., by deep-work-session).

Question Style	Batch Checkpoint Behavior
front-loaded	Skip checkpoints. Continue autonomously. Only stop if truly blocked.
minimal	Skip checkpoints. Continue autonomously. Only stop if truly blocked.
checkpoint	Pause at batch boundaries for feedback (default behavior).
collaborative	Pause at batch boundaries for feedback.
Not set	Pause at batch boundaries for feedback (default behavior).

If front-loaded or minimal: Execute all tasks continuously. Report progress but don't wait for feedback between batches. Only use the question tool if genuinely blocked.

The Process

Step 1: Load and Review Plan

1. Read plan file
2. Review critically - identify any questions or concerns about the plan
3. If concerns: Raise them with your human partner before starting
4. If no concerns: Create TodoWrite and proceed

Step 2: Execute Batch

Default batch size: 3 tasks (but execute all continuously if question style is front-loaded/minimal)

For each task:

1. Mark as in_progress
2. Follow each step exactly (plan has bite-sized steps)
3. Run verifications as specified
4. Mark as completed

Step 3: Report (or Continue)

If question style is front-loaded or minimal:

• Log progress briefly but DO NOT stop for feedback
• Continue immediately to next batch
• Only stop if genuinely blocked

Otherwise (checkpoint, collaborative, or not set):

• Show what was implemented
• Show verification output
• Say: "Ready for feedback."
• Wait for user input before continuing

Step 4: Continue

Based on feedback (if you waited for it):

• Apply changes if needed
• Execute next batch
• Repeat until complete

Step 5: Complete Development

After all tasks complete and verified:

• Announce: "I'm using the finishing-a-development-branch skill to complete this work."
• REQUIRED SUB-SKILL: Use superpowers:finishing-a-development-branch
• Follow that skill to verify tests, present options, execute choice

When to Stop and Ask for Help

STOP executing immediately when:

• Hit a blocker mid-batch (missing dependency, test fails, instruction unclear)
• Plan has critical gaps preventing starting
• You don't understand an instruction
• Verification fails repeatedly

Use the question tool to ask for clarification rather than guessing.

When to Revisit Earlier Steps

Return to Review (Step 1) when:

• Partner updates the plan based on your feedback
• Fundamental approach needs rethinking

Don't force through blockers - stop and use the question tool to ask.

Remember

• Review plan critically first
• Follow plan steps exactly
• Don't skip verifications
• Reference skills when plan says to
• Respect question style: If front-loaded/minimal, continue autonomously without asking for feedback between batches
• If question style allows checkpoints: report and wait between batches
• Stop when blocked, don't guess
• Never start implementation on main/master branch without explicit user consent

Integration

Required workflow skills:

• superpowers:using-git-worktrees - REQUIRED: Set up isolated workspace before starting
• superpowers:writing-plans - Creates the plan this skill executes
• superpowers:finishing-a-development-branch - Complete development after all tasks