executing-plans

Executes written implementation plans task-by-task with verification checkpoints and review gates.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/kxnzee/multi-repo-specs --skill executing-plans-kxnzee
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/kxnzee/multi-repo-specs/tree/main/extensions/superpowers/skills/executing-plans
Command: npx skills add https://github.com/kxnzee/multi-repo-specs --skill executing-plans-kxnzee

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Written implementation plans often drift during execution: steps get skipped, verifications are ignored, and blockers are silently worked around. This Skill enforces disciplined execution of a written plan in a separate session, with critical review before starting and explicit stop conditions when blocked. ## Core Features & Use Cases - Critical Plan Review: Loads the plan file, raises concerns with the human partner before starting, and creates todos for each plan item. - Step-by-Step Execution: Marks tasks in_progress, follows each bite-sized step exactly, runs specified verifications, and marks tasks completed. - OpenSpec Apply Mode: When invoked by OpenSpec Apply, executes only the accepted repository scope and returns verification evidence to the caller without auto-publishing or merging. - Use Case: You have a written implementation plan from a planning session. Hand it to this Skill in a fresh session so every task is executed in order, verified, and handed off to the finishing-a-development-branch workflow. ## Quick Start Execute the implementation plan in plan.md task by task, running each verification and stopping to ask if anything is blocked.

Frequently Asked Questions about executing-plans

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute an implementation plan with an AI agent?▼

Load the plan file, review it critically and raise concerns before starting, then create todos for each item. Execute each task in order, run the specified verifications, and mark tasks completed as you go.

What should an AI agent do when a plan step fails or is unclear?▼

Stop executing immediately when hitting a blocker such as a missing dependency, failing test, or unclear instruction. Ask the human partner for clarification rather than guessing or forcing through the blocker.

Does executing-plans work with OpenSpec Apply?▼

Yes. In OpenSpec Apply mode it executes only the accepted repository scope and returns verification evidence to the caller. Publication, merge, and Store Git operations remain separate authorized actions.

Can I start implementation on a protected or integration branch?▼

No. The Skill explicitly forbids starting implementation on a protected or integration branch identified by repository policy without explicit user authorization. Use an isolated workspace such as a git worktree first.

When should I use subagent-driven execution instead of inline execution?▼

Use subagent-driven development when the platform supports subagents (Claude Code, Codex CLI, Copilot CLI) and the user has authorized them. Otherwise keep the inline executor for the plan.