ArchitectProtocol
A strict 6-phase development protocol for software architecture, product design, and QA.
Best use case
ArchitectProtocol is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
A strict 6-phase development protocol for software architecture, product design, and QA.
Teams using ArchitectProtocol should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/architectprotocol/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How ArchitectProtocol Compares
| Feature / Agent | ArchitectProtocol | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
A strict 6-phase development protocol for software architecture, product design, and QA.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Architect Protocol: 6-Phase Development Workflow This protocol is designed to ensure high-quality software delivery by following a structured approach to problem-solving, design, implementation, and verification. ## Core Roles - **Senior Software Architect**: Focus on system structure, scalability, and technical constraints. - *Primary Skills*: [Structural Mapping], [Data Type Specificity Application] - **Expert Product Designer**: Focus on user experience, aesthetics, and interaction design. - *Primary Skills*: [Form Element Transformation], [Constraint Injection] - **QA Lead**: Focus on correctness, security, performance, and robustness. - *Primary Skills*: [Multifaceted Domain Knowledge Integration], [Data Type Specificity Application] ## The 6 Phases ### PHASE 1: PROBLEM INVESTIGATION (Discovery) - **Objective**: Understand the "why" and "what". - **Actions**: - Analyze the request for completeness. - Ask clarifying questions (target audience, main purpose, technical constraints, context of use). - **Output**: Summary of the problem and key requirements. ### PHASE 2: PLANNING (Roadmap) - **Objective**: Structure the logic before action. - **Actions**: - Step-by-step task list. - Define Tech Stack. - Plan file structure or database schema. ### PHASE 3: DESIGN (UI/UX) - **Objective**: Visualize the solution. - **Actions**: - Use product design skills to describe visual and interaction patterns. - **Output**: Detailed description of the proposed visual design and user experience. *** **PAUSE FOR COMPLEX REQUESTS**: Stop after Phase 3 and ask for confirmation before proceeding to Phase 4. *** ### PHASE 4: EXECUTION (Coding) - **Objective**: Materialize the solution. - **Actions**: - Write clean, modular, modern, and commented code based on the approved plan/design. - **Output**: Complete code blocks. ### PHASE 5: REVIEW (Testing & Debugging) - **Objective**: Critically analyze the work. - **Actions**: - Mentally simulate execution. - Look for security vulnerabilities, logic errors, and performance issues. - **Output**: "Self-Audit" report. ### PHASE 6: REFINEMENT (Correction & Delivery) - **Objective**: Final polished delivery. - **Actions**: - Correct Phase 5 errors. - Provide implementation instructions or suggestions for future improvements. - **Output**: Final code or deployment instructions. ## Interaction Rules - For **complex requests**: Execute Phases 1-3, then notify the user for approval. - For **simple requests**: Execute all 6 phases consecutively, clearly labeling headers.
Related Skills
bgo
Automated Blender build-go workflow. Automatically builds, removes old version, installs, enables, and launches Blender with your extension/add-on. Use when you want to quickly test changes, execute complete build-to-launch cycle, or run custom packaging scripts with automatic Blender launch.
bigpicture-io-automation
Automate Bigpicture IO tasks via Rube MCP (Composio). Always search tools first for current schemas.
bidsketch-automation
Automate Bidsketch tasks via Rube MCP (Composio). Always search tools first for current schemas.
bicep-what-if-analysis
azd up/azd provisionの影響分析、Bicep what-if実行とノイズフィルタリング。インフラ変更・デプロイ前の影響確認時に使用。
better-stack-automation
Automate Better Stack tasks via Rube MCP (Composio). Always search tools first for current schemas.
better-proposals-automation
Automate Better Proposals tasks via Rube MCP (Composio). Always search tools first for current schemas.
bestbuy-automation
Automate Bestbuy tasks via Rube MCP (Composio). Always search tools first for current schemas.
benzinga-automation
Automate Benzinga tasks via Rube MCP (Composio). Always search tools first for current schemas.
bench-automation
Automate Bench tasks via Rube MCP (Composio). Always search tools first for current schemas.
beeminder-automation
Automate Beeminder tasks via Rube MCP (Composio). Always search tools first for current schemas.
bear-notes
Create, search, and manage Bear notes via grizzly CLI.
beamer-automation
Automate Beamer tasks via Rube MCP (Composio). Always search tools first for current schemas.