Skip to content
Back to search

stream-coding

MarketplaceWorks with Claude CodeGitHub

Documentation-first development methodology. AI-ready docs = automatic code generation. 10-20x velocity.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/frmoretto--stream-coding/badge.svg)](/listings/frmoretto--stream-coding)

HTML

<a href="/listings/frmoretto--stream-coding"><img src="/api/listings/frmoretto--stream-coding/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Option 1: Claude.ai / Claude Desktop

1. Download [`stream-coding.skill`](dist/stream-coding.skill) 2. Go to Settings → Features → Skills → Add 3. Upload the `.skill` file 4. Ask Claude: *"Build a user authentication system"*

Option 2: Claude Code

Clone the repo — Claude Code auto-detects skills in `.claude/skills/`:

git clone https://github.com/frmoretto/stream-coding
cd stream-coding
# Claude Code will automatically detect .claude/skills/stream-coding/SKILL.md

Or copy `.claude/skills/stream-coding/` to your project's `.claude/skills/` directory.

Ask Claude: *"Build a user authentication system"*

Option 3: Claude Projects

Add [`SKILL.md`](SKILL.md) to project knowledge. Claude will search it when needed, though Skills provide better integration.

Option 4: OpenAI Codex / GitHub Copilot

Copy the canonical skill to the appropriate directory:

| Platform | Location | |----------|----------| | OpenAI Codex | `.codex/skills/stream-coding/SKILL.md` | | GitHub Copilot | `.github/skills/stream-coding/SKILL.md` |

Use [`skills/stream-coding/SKILL.md`](skills/stream-coding/SKILL.md) (agentskills.io format).

Option 5: Manual / Other LLMs

For Cursor, Windsurf, or other AI tools: - Extract core principles (Phases, Document Types, Spec Gate) - Create a condensed version for `.cursorrules` or project settings - Use the [templates](./templates) and [Spec Gate Checklist](./templates/SPEC_GATE_CHECKLIST.md) as reference

The methodology is tool-agnostic—only SKILL.md is Claude-optimized.

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 38 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "$schema": "https://claude.ai/schemas/plugin-manifest-v1.json",
  3. "name": "stream-coding",
  4. "version": "3.4.0",
  5. "description": "Documentation-first development methodology. AI-ready docs = automatic code generation. 10-20x velocity.",
  6. "author": "Francesco Marinoni Moretto",
  7. "license": "CC-BY-4.0",
  8. "repository": "https://github.com/frmoretto/stream-coding",
  9. "homepage": "https://github.com/frmoretto/stream-coding",
  10. "keywords": [
  11. "stream-coding",
  12. "documentation-first",
  13. "methodology",
  14. "clarity-gate",
  15. "ai-ready",
  16. "specs",
  17. "10x-velocity"
  18. ],
  19. "skills": [
  20. {
  21. "name": "stream-coding",
  22. "path": "skills/stream-coding/SKILL.md",
  23. "triggers": [
  24. "stream coding",
  25. "documentation-first",
  26. "build",
  27. "create",
  28. "implement",
  29. "document",
  30. "spec out"
  31. ]
  32. }
  33. ],
  34. "compatibility": {
  35. "claude-code": ">=1.0.0",
  36. "claude-desktop": ">=1.0.0"
  37. }
  38. }
README.md · README · 249 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Stream Coding v3.5
  2. **The 10-20x Methodology for AI-Accelerated Software Development**
  3. [![License: CC BY 4.0](https://img.shields.io/badge/License-CC_BY_4.0-lightgrey.svg)](https://creativecommons.org/licenses/by/4.0/)
  4. > 📢 **v3.5 Update:** Phase 2.5 Adversarial Review added, internal gate renamed Spec Gate (structural), Clarity Gate is now a standalone epistemic tool. Re-download if using older version.
  5. > ✅ **This methodology built 7 production modules in 4.5 hours** ([5Levels Case Study](./manifesto/Appendix_C_5Levels_Case_Study.md), Git-verified)
  6. > *"Stream coding isn't about faster coding. It's about documentation so clear that code writes itself."*
  7. ---
  8. ## The Problem: The Velocity Mirage
  9. AI tools promise 10x productivity. GitHub Copilot, Cursor, Claude Code—they make coding 55% faster.
  10. **But projects still take the same time to ship.**
  11. Why? Because faster typing doesn't solve:
  12. - Strategic decisions AI can't make for you
  13. - Context that gets lost between prompts
  14. - Technical debt created at 10x speed
  15. This gap between task velocity and project velocity is the **Velocity Mirage**.
  16. ---
  17. ## The Solution: Stream Coding
  18. Stream Coding is a documentation-first methodology that makes AI-generated code deterministic.
  19. **The 40/40/5/10/5 Split:**
  20. - **40%** Strategic Thinking (Phase 1) — Solve hard problems before coding
  21. - **40%** AI-Ready Documentation (Phase 2) — Specs so complete AI has zero decisions
  22. - **5%** Adversarial Review (Phase 2.5) — Different AI attacks the specs before coding
  23. - **10%** Execution (Phase 3) — Code streams out automatically
  24. - **5%** Quality (Phase 4) — Tests and verification
  25. **Real Results (5Levels Case Study):**
  26. - 7 production modules in 4.5 hours
  27. - 46 intelligence endpoints (77 total backend API)
  28. - Zero bugs in generated code, 21 minutes average per tested module
  29. *Note: The case study focuses on backend intelligence modules—Stream Coding's sweet spot. For frontend, use the methodology for behavior (components, state, logic) and complement with AI design tools for visuals. See Chapter 4 for details.*
  30. ---
  31. ## Quick Start
  32. ### Option 1: Claude.ai / Claude Desktop
  33. 1. Download [`stream-coding.skill`](dist/stream-coding.skill)
  34. 2. Go to Settings → Features → Skills → Add
  35. 3. Upload the `.skill` file
  36. 4. Ask Claude: *"Build a user authentication system"*
  37. ### Option 2: Claude Code
  38. Clone the repo — Claude Code auto-detects skills in `.claude/skills/`:
  39. ```bash
  40. git clone https://github.com/frmoretto/stream-coding
  41. cd stream-coding
  42. # Claude Code will automatically detect .claude/skills/stream-coding/SKILL.md
  43. ```
  44. Or copy `.claude/skills/stream-coding/` to your project's `.claude/skills/` directory.
  45. Ask Claude: *"Build a user authentication system"*
  46. ### Option 3: Claude Projects
  47. Add [`SKILL.md`](SKILL.md) to project knowledge. Claude will search it when needed, though Skills provide better integration.
  48. ### Option 4: OpenAI Codex / GitHub Copilot
  49. Copy the canonical skill to the appropriate directory:
  50. | Platform | Location |
  51. |----------|----------|
  52. | OpenAI Codex | `.codex/skills/stream-coding/SKILL.md` |
  53. | GitHub Copilot | `.github/skills/stream-coding/SKILL.md` |
  54. Use [`skills/stream-coding/SKILL.md`](skills/stream-coding/SKILL.md) (agentskills.io format).
  55. ### Option 5: Manual / Other LLMs
  56. For Cursor, Windsurf, or other AI tools:
  57. - Extract core principles (Phases, Document Types, Spec Gate)
  58. - Create a condensed version for `.cursorrules` or project settings
  59. - Use the [templates](./templates) and [Spec Gate Checklist](./templates/SPEC_GATE_CHECKLIST.md) as reference
  60. The methodology is tool-agnostic—only SKILL.md is Claude-optimized.
  61. ---
  62. ## Read the Manifesto
  63. The [`/manifesto`](./manifesto) folder contains the complete methodology:
  64. | Chapter | Topic |
  65. |---------|-------|
  66. | [Chapter 1](./manifesto/Chapter_01_The_10x_Promise.md) | The Velocity Mirage |
  67. | [Chapter 2](./manifesto/Chapter_02_Why_AI_Cant_Deliver.md) | Why AI Tools Alone Fail |
  68. | [Chapter 3](./manifesto/Chapter_03_The_Methodology_Gap.md) | The Missing Middle |
  69. | [Chapter 4](./manifesto/Chapter_04_What_Is_Stream_Coding.md) | The 5-Phase Methodology |
  70. | [Chapter 5](./manifesto/Chapter_05_The_Perpetual_Stream.md) | Day 2 & The Rule of Divergence |
  71. | [Appendix A](./manifesto/Appendix_A_The_Toolkit.md) | Templates & Checklists |
  72. | [Appendix B](./manifesto/Appendix_B_References.md) | Research & SDD Positioning |
  73. | [Appendix C](./manifesto/Appendix_C_5Levels_Case_Study.md) | 5Levels Case Study (Git-Verified) |
  74. | [Advanced Framework](./manifesto/advanced/Advanced_Framework.md) | Document Architecture (v3.5) |
  75. ### Use the Templates
  76. The [`/templates`](./templates) folder contains ready-to-use frameworks:
  77. | Template | Purpose |
  78. |----------|--------|
  79. | [Strategic Blueprint](./templates/STRATEGIC_BLUEPRINT.md) | Answer the 7 Phase 1 Questions |
  80. | [ADR Template](./templates/ADR_TEMPLATE.md) | Document architecture decisions with rationale |
  81. | [Spec Gate Checklist](./templates/SPEC_GATE_CHECKLIST.md) | The mandatory Phase 2→2.5 gate |
  82. ---
  83. ## The Core Insight
  84. > **"When code fails, fix the spec—not the code."**
  85. Traditional development iterates on code. Stream Coding iterates on documentation.
  86. Every manual code edit without updating the spec creates **Divergence**—technical debt that breaks the stream. The methodology works because it treats code as a compiled output of documentation, not the source of truth.
  87. ---
  88. ## Who This Is For
  89. - ✅ **Technical founders** building greenfield products
  90. - ✅ **Solo developers** and small teams (1-5 people)
  91. - ✅ **Anyone tired of AI-generated spaghetti code**
  92. - ✅ **Backend/business logic focused** (see Chapter 4 for frontend approach)
  93. - ❌ Not for large enterprises (see GitHub Spec-Kit, Kiro, Gemini Conductor)
  94. - ❌ Not for teams who can't commit to documentation-first
  95. ---
  96. ## Where Stream Coding Fits
  97. ```
  98. ┌─────────────────────────────────────────────────────────────────┐
  99. │ AI-Assisted Development Landscape │
  100. ├─────────────────────────────────────────────────────────────────┤
  101. │ │
  102. │ Layer 3: Enterprise SDD GitHub Spec-Kit, Kiro, Conductor │
  103. │ (teams 50+) Full lifecycle, heavy process │
  104. │ │
  105. │ Layer 2: Founder SDD ◄── STREAM CODING │
  106. │ (teams 1-5) Documentation-first, AI-ready specs │
  107. │ │
  108. │ Layer 1: AI Assistants Copilot, Cursor, Claude Code │
  109. │ (task-level) Fast typing, no methodology │
  110. │ │
  111. │ Layer 0: Manual Coding Traditional development │
  112. │ │
  113. └─────────────────────────────────────────────────────────────────┘
  114. ```
  115. **The Gap:** Layer 1 tools make coding 55% faster but don't solve the methodology problem. Enterprise tools (Layer 3) are overkill for founders. Stream Coding fills the middle.
  116. ---
  117. ## Competitive Positioning
  118. | Dimension | AI Assistants (Copilot) | Stream Coding | Enterprise SDD (Kiro) |
  119. |-----------|------------------------|---------------|----------------------|
  120. | **Target** | Individual developers | Founders, small teams | Large enterprises |
  121. | **Focus** | Task velocity | Project velocity | Process compliance |
  122. | **Cost** | $10-20/mo | Free | Enterprise pricing |
  123. | **Methodology** | None | 5-phase, Spec Gate | Full SDLC |
  124. | **Documentation** | Optional | Mandatory (80% of time) | Mandatory |
  125. | **AI Role** | Code suggestions | Spec execution | Workflow automation |
  126. ---
  127. ## Research Validation
  128. Stream Coding aligns with industry research:
  129. - **McKinsey (2025):** Top performers see 16-30% productivity gains through "end-to-end PDLC implementation" and "structured communication of specs"
  130. - **DORA (2025):** 7.2% delivery instability increase for every 25% AI adoption *without foundational systems*
  131. - **METR (2025):** Developers 19% slower with AI despite *feeling* 20% faster
  132. The methodology isn't magic, it's systematic application of spec-driven development at founder scale.
  133. ---
  134. ## Related
  135. **Clarity Gate** — Pre-ingestion verification for epistemic quality (originated from Stream Coding's Clarity Gate concept)
  136. [github.com/frmoretto/clarity-gate](https://github.com/frmoretto/clarity-gate)
  137. **Source of Truth Creator** — Create epistemically calibrated documents
  138. [github.com/frmoretto/source-of-truth-creator](https://github.com/frmoretto/source-of-truth-creator)
  139. ---
  140. ## Roadmap
  141. | Phase | Status | Description |
  142. |-------|--------|-------------|
  143. | **v3.4** | ✅ Released | 13-item Clarity Gate, scoring rubric, Documentation Audit |
  144. | **v3.5** | ✅ Released | Phase 2.5 Adversarial Review, Spec Gate (structural), Clarity Gate (epistemic) split |
  145. | **v4.0** | 🔜 Planned | Multi-agent workflow support, automated spec validation |
  146. ---
  147. ## Contributing
  148. Stream Coding is open source under CC BY 4.0. You're free to use, adapt, and share with attribution.
  149. Looking for:
  150. 1. **Real-world case studies** — Applied Stream Coding to your project?
  151. 2. **Tool integrations** — Cursor, Windsurf, other AI tools
  152. 3. **Methodology feedback** — Are the 5 phases the right split?
  153. 4. **Template improvements** — Better Strategic Blueprint, ADR formats
  154. Open an issue or PR.
  155. ---
  156. ## About
  157. Created by [Francesco Marinoni Moretto](https://www.linkedin.com/in/francesco-moretto/) while building [5Levels](https://5levels.io), a LinkedIn relationship intelligence platform.
  158. The methodology emerged from building 7 production modules in 4.5 hours, and documenting exactly how.
  159. ---
  160. ## License
  161. [CC BY 4.0](./LICENSE) — Use freely with attribution.
  162. *"Stream Coding methodology by Francesco Marinoni Moretto (github.com/frmoretto/stream-coding)"*
  163. ---
  164. *In memory of my beloved father Guido*

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.