executing-plans

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

Updated May 22, 2026
One-click install
npx skills add https://github.com/viniciuscs84/sdd-toolkit --skill executing-plans-viniciuscs84
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/viniciuscs84/sdd-toolkit/tree/main/skills/executing-plans%20copy
Command: npx skills add https://github.com/viniciuscs84/sdd-toolkit --skill executing-plans-viniciuscs84

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Written implementation plans often get executed inconsistently—steps are skipped, verifications are ignored, and blockers are worked around with guesses. This Skill enforces a disciplined process for loading a plan, reviewing it critically, executing every task in order, and stopping to ask for help instead of improvising. ## Core Features & Use Cases - Critical Plan Review: Loads the plan file and raises questions or concerns with the human partner before any code is written. - Task-by-Task Execution: Tracks each task with a todo list, follows bite-sized steps exactly, and runs the verifications specified in the plan. - Stop-and-Ask Guardrails: Halts execution on blockers, unclear instructions, or repeated verification failures instead of guessing. - Use Case: You have a multi-task implementation plan written in a previous session. Hand it to the agent, which reviews it, executes each task with verification, and finishes the development branch with tests and review options. ## Quick Start Use the executing-plans skill to implement the plan in docs/plans/feature-auth.md and stop if anything is unclear.

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?

Provide the plan file and invoke the executing-plans skill. The agent reviews the plan critically, raises concerns before starting, then executes each task in order using a todo list, running the verifications specified in the plan.

What should an AI agent do when a plan step fails?

The agent should stop executing immediately when it hits a blocker, a failing test, or an unclear instruction. It asks the human partner for clarification rather than guessing or forcing through the failure.

Can I use executing-plans without subagent support?

Yes, this skill is designed for sessions without subagents. If your platform supports subagents, such as Claude Code or Codex, the documentation recommends using subagent-driven development instead for higher quality results.

What happens after all plan tasks are completed?

After all tasks are verified, the agent invokes the finishing-a-development-branch skill. That skill verifies tests, presents completion options, and executes the chosen path to close out the work.

Does executing-plans work on the main branch?

No. The skill explicitly forbids starting implementation on the main or master branch without explicit user consent, and it integrates with a git worktree skill to ensure an isolated workspace.