spec-writing

Generate structured technical specifications with RFC 2119 terminology.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/pjordan/claude-toolkit --skill spec-writing-pjordan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-writing
Source: https://github.com/pjordan/claude-toolkit/tree/main/skills/examples/spec-writing
Command: npx skills add https://github.com/pjordan/claude-toolkit --skill spec-writing-pjordan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Spec Writing Skill helps teams convert vague ideas into precise, testable technical specifications, reducing ambiguity and rework.

Core Features & Use Cases

  • Structured templates for feature, API, architecture, and data specs
  • RFC-compliant language with RFC 2119 keywords
  • Guided workflow from context gathering to acceptance criteria
  • Edge cases, non-functional requirements, and testing considerations

Quick Start

Provide a brief context about the problem, goals, and constraints. Then ask Claude to draft a complete spec using the appropriate template. Review the generated draft, iteratively refine requirements, data models, and acceptance criteria, and finalize the document.

Frequently Asked Questions about spec-writing

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

FAQPage Schema
How do I write a technical specification from a vague feature idea?

To write a technical specification from a vague feature idea, provide the problem context, goals, and constraints to trigger automated drafting. The process uses structured templates for features, APIs, architecture, and data models, enforcing RFC 2119 terminology and acceptance criteria to reduce ambiguity and rework.

What is the best way to structure API design requirements and acceptance criteria?

The best way to structure API design requirements is using guided templates that enforce RFC 2119 keywords for compliance language. This approach systematically captures data models, edge cases, non-functional requirements, and specific acceptance criteria, ensuring the API specification remains testable and precise across software development projects.

How do I create an RFC document with correct terminology for architecture decisions?

To create an RFC document for architecture decisions, use a structured specification template that enforces RFC 2119 terminology. This ensures precise language for requirements and constraints, guiding the workflow from initial context gathering through final acceptance criteria for technical architecture choices.

Can I generate testable data models and non-functional requirements from a brief description?

Yes, you can generate testable data models and non-functional requirements from a brief description. By providing context about your problem and constraints, the specification drafting process automatically structures these elements into a precise template, capturing edge cases and testing considerations alongside the data model.

What should I include in a technical specification to prevent project rework?

To prevent project rework, a technical specification should include structured sections for versioning, edge cases, non-functional requirements, and explicit acceptance criteria. Using RFC 2119 keywords ensures precise language, converting ambiguous ideas into testable requirements that reduce development friction.

Does the specification template support versioning for iterative feature planning?

Yes, the specification template supports versioning for iterative feature planning. You can draft a complete spec using the appropriate template, review the generated output, and iteratively refine requirements, data models, and acceptance criteria as the feature or API design evolves.