executing-plans

Executes written implementation plans in batches with review checkpoints between task groups.

Updated Mar 24, 2026
One-click install
npx skills add https://github.com/AarushiShah/coding-agents-databricks-apps --skill executing-plans-aarushishah
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/AarushiShah/coding-agents-databricks-apps/tree/main/.claude/skills/executing-plans
Command: npx skills add https://github.com/AarushiShah/coding-agents-databricks-apps --skill executing-plans-aarushishah

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Executing a multi-task implementation plan in one shot often leads to drift, skipped verifications, and undetected errors. This Skill enforces a disciplined workflow where plans are reviewed critically, executed in small batches, and paused for human feedback between batches. ## Core Features & Use Cases - Critical Plan Review: Loads the plan file and raises questions or concerns with the human partner before writing any code. - Batched Execution with Checkpoints: Executes tasks in batches (default 3), marking progress with TodoWrite and running specified verifications per task. - Stop-and-Ask Discipline: Halts immediately on blockers, unclear instructions, or repeated verification failures instead of guessing. - Use Case: You have a 12-task refactoring plan written by an architect. Use this Skill to implement the first three tasks, report verification output, wait for feedback, then continue batch by batch until completion. ## Quick Start Use the executing-plans skill to implement the plan in docs/refactor-plan.md, executing three tasks at a time and checking in with me between batches.

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 coding agent?

Load the plan file, review it critically for gaps or concerns, then execute tasks in batches of three with verification after each task. Report results between batches and wait for feedback before continuing to the next batch.

How many tasks should be executed before checking in for review?

The default batch size is three tasks. After completing a batch with all verifications passing, report what was implemented, show verification output, and wait for feedback before starting the next batch.

What should I do when a plan task fails verification?

Stop executing immediately when verification fails repeatedly, a dependency is missing, or an instruction is unclear. Ask the human partner for clarification rather than guessing or forcing through the blocker.

When should a plan be sent back for revision instead of executed?

Return to plan review when the plan has critical gaps preventing a start, when the fundamental approach needs rethinking, or when the partner updates the plan based on raised concerns. Never force execution through a flawed plan.

What happens after all plan tasks are completed?

After all tasks complete and pass verification, the workflow hands off to the finishing-a-development-branch skill, which verifies tests, presents completion options, and executes the chosen finalization step.