skills.rest
nuttaruj
nuttarujCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

implement-plan

Execute approved implementation plans with risk-based testing and structured manifests.

3|1|Updated May 10, 2026
Search Tags:#workflow#tdd#implementation#verification#subagents#delegation#code editing
One-click install
npx skills add https://github.com/nuttaruj/rolepod --skill implement-plan-nuttaruj
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: implement-plan
Source: https://github.com/nuttaruj/rolepod/tree/main/core/skills/implement-plan
Command: npx skills add https://github.com/nuttaruj/rolepod --skill implement-plan-nuttaruj

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you execute approved implementation work without losing scope, skipping verification, or making risky code changes casually. It provides a disciplined build workflow so code edits stay aligned to the plan, use the right level of testing, and hand off cleanly to verification.

Core Features & Use Cases

  • Plan-driven execution: Reads the approved plan, verifies touched files first, and treats the plan's named commands as the source of truth for task completion.
  • Risk-based testing discipline: Enforces test-first work for bug fixes and high-risk areas like auth, billing, migrations, security, and concurrency, while allowing lighter evidence for low-risk edits.
  • Bounded delegation and review: Supports tightly scoped subagent tasks, structured manifests, fresh-context reviews, and whole-implementation review before handoff.
  • Use cases: Implementing a multi-step feature from a written plan, making a surgical single-file fix, coordinating parallel task execution, or reviewing delegated code changes before moving to verification.

Quick Start

Ask the AI to use implement-plan to execute your approved task list, read the target files first, make only the planned edits, and verify each task with its specified command.

Frequently Asked Questions about implement-plan

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

FAQPage Schema
How do I execute an approved implementation plan without losing scope?▼

Risk-based testing enforces test-first workflows for bug fixes and high-risk areas like auth, billing, or concurrency, while allowing lighter evidence for low-risk edits. This ensures critical paths are verified before integration.

Can I delegate subagent work for parallel development tasks?▼

Yes, you can implement a surgical single-file fix by treating the plan's named commands as the source of truth for task completion. The workflow reads target files first, makes only the planned edits, and verifies the task with its specified command.

What is the best way to verify code changes before moving to verification?▼

The best way to verify code changes is producing a structured implementation manifest and conducting a whole-implementation review with fresh context. This ensures the executed work matches the plan before handing off to the verification phase.

When do I need test-first workflow for code editing?▼

You need a test-first workflow for code editing when touching high-risk areas such as auth, billing, migrations, security, and concurrency, or when applying bug fixes. Low-risk edits allow lighter testing evidence.

Related Skills

Explore More Software Engineering Skills→
daffy0208daffy0208
34

BMAD Method

Map business models to technical architecture for scalable infrastructure planning.

Community
Intermediate
HigorAlvesHigorAlves
4

verification-before-completion

Run verification commands and review outputs before confirming task completion.

Community
Basic
BlackRoad-OS-IncBlackRoad-OS-Inc
1

subagent-driven-development

Distribute plan tasks to fresh subagents with isolated contexts and two-stage reviews.

Official
Advanced
rhuangabrielsantosrhuangabrielsantos

cy-review-round

Review PRD implementation code and generate structured issue files.

Community
Intermediate