unikit-implement

Execute feature-plan tasks from .unikit/plans with optional parallel agents.

16|1|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/NintendaDev/unikit-ai --skill unikit-implement
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: unikit-implement
Source: https://github.com/NintendaDev/unikit-ai/tree/main/skills/unikit-implement
Command: npx skills add https://github.com/NintendaDev/unikit-ai --skill unikit-implement

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Executes end-to-end implementation tasks from a feature plan stored in .unikit/plans/, reading TASKS.md and other plan notes, bootstrapping project rules and applying them to implement tasks either inline or via subagents as needed.

Core Features & Use Cases

  • Reads .unikit/plans/ folders to identify pending tasks and supports Phase/Task selectors for targeted execution.
  • Boots rules and principles once at start to guide all task implementations, ensuring consistency with project-wide standards.
  • Executes tasks inline when tasks are tightly coupled or require fast iteration, or spawns parallel agents for independent work scopes, with automatic plan updates.
  • Updates TASKS.md to reflect completed work, optionally runs tests, creates documentation checkpoints, and prepares a completion summary.

Quick Start

Start by loading the latest feature plan from .unikit/plans and execute the next pending task.

Frequently Asked Questions about unikit-implement

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

FAQPage Schema
How do I execute feature-plan tasks end-to-end for software projects?

To execute feature-plan tasks end-to-end, load the feature plan from your .unikit/plans/ directory and the system will read TASKS.md to implement pending items inline or via parallel agents. It automatically applies bootstrapped project rules to ensure consistency.

What is the best way to automate task implementation from a feature plan?

Automating task implementation from a feature plan is best handled by using Phase and Task selectors to target specific pending work. The system applies bootstrapped rules, implements tasks inline or via parallel agents, and updates TASKS.md with completed progress.

How does parallel agent execution work for independent feature-plan tasks?

Parallel agent execution for independent feature-plan tasks works by spawning subagents for independent work scopes, while executing tightly coupled tasks inline. This allows simultaneous implementation of separate tasks before updating the plan and preparing a completion summary.

Do I need a specific folder structure to automate task implementation?

Yes, you need a specific folder structure to automate task implementation. Your software project must store feature plans and notes under the .unikit/plans/ directory, including a TASKS.md file to determine the pending work scope and track completion.

Can I run tests and create documentation checkpoints after implementing tasks?

You can run tests and create documentation checkpoints after implementing tasks. Once tasks are complete, the system optionally runs tests, creates documentation checkpoints, and prepares a completion summary to reflect the finished work.