skills.rest
awslabs
awslabsOFFICIAL
Explore all skills by this author →
WebsiteGitHub

use-case-specification

Creates a use case specification document defining business problems, stakeholders, and success criteria for model customization.

881|152|Updated Feb 5, 2026
Search Tags:#fine-tuning#requirements gathering#success criteria#responsible ai#model customization#Sagemaker#use case specification
One-click install
npx skills add https://github.com/awslabs/agent-plugins --skill use-case-specification
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: use-case-specification
Source: https://github.com/awslabs/agent-plugins/tree/main/plugins/sagemaker-ai/skills/use-case-specification
Command: npx skills add https://github.com/awslabs/agent-plugins --skill use-case-specification

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Teams starting model customization often skip defining the business problem and measurable success criteria, leading to unfocused fine-tuning efforts. This Skill guides a structured conversation to produce a reusable use case specification aligned with the AWS Responsible AI Lens.

Core Features & Use Cases

  • Guided Discovery: Collects the problem statement, primary users, and LLM-as-a-Judge success criteria through a short multi-turn conversation, inferring details from context instead of interrogating.
  • Specification Generation: Produces a Markdown file named [relevant_title]_use_case_spec.md containing Business Problem, Primary Users, and Success Tenets fields.
  • Reuse and Edit Protocol: Detects existing use case spec files for reuse and supports iterative edits with user confirmation.
  • Use Case: Before fine-tuning a model on SageMaker AI, run this Skill to capture why the customization is needed, who will use the model, and how success will be measured.

Quick Start

Ask the agent to create a use case specification for your planned model customization project.

Frequently Asked Questions about use-case-specification

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

FAQPage Schema
How do I create a use case specification for model customization?▼

Start the skill and answer a few discovery questions about the problem, users, and success criteria. It infers details from your conversation and generates a Markdown file named [relevant_title]_use_case_spec.md for your approval.

What is a use case specification in the AWS Responsible AI Lens?▼

A use case specification is a design principle recommended by the AWS Responsible AI Lens that documents the business problem, primary users, and success tenets before model customization. It ensures customization efforts are grounded in measurable outcomes.

What success criteria work for evaluating a fine-tuned model?▼

Success criteria must be measurable by an LLM-as-a-Judge, such as response accuracy or tone adherence. Infrastructure metrics like latency or throughput are not valid success tenets for this specification.

Can I reuse or edit an existing use case specification file?▼

Yes. The skill checks for existing *_use_case_spec.md files and asks whether to reuse, modify, or start fresh. You can also edit the file directly, and the skill will read the updated version.

Does the use case specification skill handle base model selection?▼

No. Base model selection is explicitly out of scope and handled by a separate model-selection skill. This skill focuses only on the problem statement, stakeholders, and success criteria.

Related Skills

Explore More Product & Management Skills→
MasterleeausMasterleeaus

aiox-pm

Create product requirements documents, epics, and strategic product artifacts.

Community
Advanced
mahg-esmahg-es
3

model-fine-tuning

Fine-tune open-source LLMs with LoRA/QLoRA adapters and prepared datasets.

Official
Advanced
oldwinteroldwinter
5

ai-evals

Design AI/LLM feature evaluation plans with test sets, rubrics, and judging approaches.

Community
Advanced
JaviMontanoJaviMontano

metodologia-aws-architecture-design

Design AWS GenAI architectures for scalable, secure AI workloads.

Community
Advanced