gsd-ai-integration-phase

Generates AI-SPEC design contracts for AI development phases with evaluation strategy.

1|Updated Sep 6, 2015
One-click install
npx skills add https://github.com/deerawan/dotfiles --skill gsd-ai-integration-phase-deerawan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gsd-ai-integration-phase
Source: https://github.com/deerawan/dotfiles/tree/main/claude/skills/gsd-ai-integration-phase
Command: npx skills add https://github.com/deerawan/dotfiles --skill gsd-ai-integration-phase-deerawan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps you quickly convert an AI development phase concept into a clear, executable design contract that defines research inputs and evaluation strategy.

Core Features & Use Cases

  • Framework-first AI integration planning: Selects an AI framework as the starting point for the phase’s design.
  • Targeted research workflow: Pulls relevant AI framework and eval material, then drills into the specific domain for the phase.
  • Eval strategy design: Produces an evaluation plan that makes the phase’s success criteria explicit and testable.

Use it when you’re planning an AI-building milestone that requires both domain understanding and a concrete evaluation approach.

Quick Start

Run the skill and pass the phase number you want to plan.

Frequently Asked Questions about gsd-ai-integration-phase

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

FAQPage Schema
How do I create an eval-ready AI design contract for a specific development phase?

Generate an eval-ready AI design contract by orchestrating framework selection, research, and evaluation planning for a specified phase number. The workflow reads reference docs and writes a contract preserving workflow gates.

What is an AI-SPEC design contract and when do I need one for workflow orchestration?

An AI-SPEC design contract defines research inputs and evaluation strategy for an AI development phase. You need one during planning and documentation workflows to make phase success criteria explicit, testable, and executable.

How do I design an evaluation strategy for an AI development phase?

Design an evaluation strategy by pulling relevant AI framework material, drilling into the specific domain, and producing an evaluation plan that makes phase success criteria explicit and testable.

Can I select an AI framework before doing domain research for my development phase?

Yes, the workflow uses a framework-first approach, selecting an AI framework as the starting point before pulling relevant material and drilling into the specific domain for the phase.

Do I need to provide a phase number to generate the AI-SPEC contract?

Yes, you must pass the phase number you want to plan. The workflow requires this input to generate an eval-ready contract that fits the specified phase and preserves the configured workflow gates.

What are the limitations of using workflow orchestration for AI integration planning?

The workflow requires reading reference docs from configured context paths to write the contract. If context paths are not configured or accessible, it cannot generate the eval-ready design contract for the selected phase.