Where are the specifications stored?

Specifications are generated as Markdown files within the docs/Features/ directory of your repository, ensuring they are version-controlled alongside your code.

Does it support visual diagrams?

Yes, the skill creates and updates Mermaid diagrams to visualize module interactions and primary user flows within your documentation.

How does it help with AI coding agents?

It provides a structured, 'anti-guesswork' specification that gives AI agents the exact rules, modules, and flows they need to implement code correctly without hallucinations.

What is the benefit of the 'Definition of Done' in this skill?

It enforces a strict quality standard including automated test coverage, clean static analysis, and updated architectural maps before a feature is considered complete.

What is the mcaf-feature-spec skill?

It is a specialized capability for Claude Code that helps users create and maintain high-quality feature specifications within the MCAF framework.

MCAF Feature Specification

Name: MCAF Feature Specification
Author: managedcode

bymanagedcode

•

Learning & Documentation

Generates and updates high-precision feature specifications with Mermaid diagrams and executable verification steps to eliminate implementation guesswork.

The mcaf-feature-spec skill automates the creation of comprehensive technical documentation within the Managed Code AI Framework (MCAF). It guides the user through defining business rules, user flows, and system behaviors using a standardized template, ensuring that every feature has a clear implementation plan and a verifiable 'Definition of Done'. By integrating Mermaid diagrams and traceability to tests, it provides a single source of truth that allows both human developers and AI coding agents to build software without ambiguity or architectural drift.

Key Features

01Automated Mermaid diagram generation for visual system flows

02Tight integration with Architectural Decision Records (ADRs) and system maps

03Executable verification plans with direct traceability to automated tests

04Anti-guesswork checklist to ensure all business rules are testable and concrete

05Standardized feature documentation using the Feature-Template.md

0632 GitHub stars

Use Cases

01Updating existing documentation when system behavior changes to maintain a single source of truth

02Defining a complex new feature before starting the implementation phase to ensure alignment

03Providing AI coding agents with the necessary context and constraints to generate production-ready code

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add managedcode/mcaf mcaf-feature-spec

For use in Claude.ai and ChatGPT

Download Skill