executing-plans
💡 Summary
A skill for AI agents to execute pre-written implementation plans in batches with human review checkpoints.
🎯 Target Audience
🤖 AI Roast: “It's a glorified to-do list manager that finally teaches AI agents to ask for directions instead of confidently driving into a ditch.”
The skill orchestrates execution of arbitrary tasks from a plan file, which is a high-risk operation. It likely invokes other skills/tools (e.g., shell, file ops) without intrinsic validation of the plan's commands. Mitigation: Implement a plan validation and sanitization step (e.g., allow-list of permitted operations) before any execution.
name: executing-plans description: Use when you have a written implementation plan to execute in a separate session with review checkpoints
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."
The Process
Step 1: Load and Review Plan
- Read plan file
- Review critically - identify any questions or concerns about the plan
- If concerns: Raise them with your human partner before starting
- If no concerns: Create TodoWrite and proceed
Step 2: Execute Batch
Default: First 3 tasks
For each task:
- Mark as in_progress
- Follow each step exactly (plan has bite-sized steps)
- Run verifications as specified
- Mark as completed
Step 3: Report
When batch complete:
- Show what was implemented
- Show verification output
- Say: "Ready for feedback."
Step 4: Continue
Based on feedback:
- 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
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 ask.
Remember
- Review plan critically first
- Follow plan steps exactly
- Don't skip verifications
- Reference skills when plan says to
- Between batches: just report and wait
- Stop when blocked, don't guess
Pros
- Enforces structured, reviewable execution reducing errors.
- Promotes safe batching and clear human-AI collaboration.
- Improves reliability by mandating verification steps.
Cons
- Heavily dependent on quality of the pre-written plan.
- Rigid batch size (3 tasks) may not suit all contexts.
- Adds overhead for simple or well-understood tasks.
Related Skills
mcp-builder
S“This guide is so comprehensive it might just teach the AI to write its own MCP servers and put you out of a job.”
learn-claude-code
A“A tutorial that finally admits its own past mistakes is more honest than most, but still can't resist the classic 'one weird loop' oversimplification.”
connect
A“It's the ultimate 'I'll do it for you' skill, turning Claude from a thoughtful advisor into an over-eager intern with access to all your accounts.”
Disclaimer: This content is sourced from GitHub open source projects for display and rating purposes only.
Copyright belongs to the original author obra.