stream-coding

Generates production-grade code from AI-ready documentation using a four-phase workflow and Clarity Gate scoring rubric.

97|8|Updated Nov 24, 2025
One-click install
npx skills add https://github.com/frmoretto/stream-coding --skill stream-coding
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stream-coding
Source: https://github.com/frmoretto/stream-coding/tree/main
Command: npx skills add https://github.com/frmoretto/stream-coding --skill stream-coding

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Stream Coding addresses the velocity Mirage by making AI-ready documentation the primary artifact, enabling production-grade code to be generated automatically from precise specifications.

Core Features & Use Cases

  • Documentation-first methodology that spans Phase 1 (Strategic Thinking), Phase 2 (AI-Ready Documentation), and Phase 3-4 (Execution and Quality)
  • 13-item Clarity Gate with scoring rubric and self-assessment for AI-coder understandability
  • Tool-agnostic workflow compatible with Claude, GPT-based assistants, Cursor, Windsurf, and others
  • Deep-links, anti-patterns, test cases, and error handling embedded in Implementation Docs to prevent divergence
  • Optimized for founder-scale teams (1-5 people) shipping production-grade modules quickly

Quick Start

Begin Phase 1 by answering the seven Phase 1 questions and creating a Strategic Blueprint before writing any code.

Frequently Asked Questions about stream-coding

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

FAQPage Schema
How do I generate production-ready code from AI documentation specifications?

You generate production-ready code by first translating structured documentation into AI-ready specs and passing a mandatory Clarity Gate. This documentation-first methodology ensures AI coders receive precise specifications before generating the final codebase.

What is the Clarity Gate in AI-assisted code generation?

The Clarity Gate is a 13-item scoring rubric with self-assessment to ensure AI-coder understandability. It acts as a mandatory checkpoint before code generation, verifying that your documentation is completely clear for AI execution.

How do I stop AI code generation from diverging from my project specifications?

To stop AI code divergence, embed deep-links, anti-patterns, test cases, and error handling directly within your Implementation Docs. This documentation-first approach provides strict boundaries for AI assistants during execution.

Can I use a documentation-first AI workflow with code editors like Cursor or Windsurf?

Yes, this workflow is tool-agnostic and compatible with Claude, GPT-based assistants, Cursor, and Windsurf. The methodology relies on generating precise documentation that any standard AI coding tool can process.

What is the best way for founder-scale teams to maintain code generation velocity?

Founder-scale teams maintain velocity by adopting a four-phase workflow spanning Strategic Thinking, AI-Ready Documentation, Execution, and Quality. This methodology makes documentation the primary artifact to automate production-grade code generation.

How do I start a documentation-first AI coding workflow?

You start by entering Phase 1: Strategic Thinking, answering seven foundational questions to create a Strategic Blueprint. You must complete this documentation phase before writing any code to ensure AI readiness.