architecture

Guides interactive creation of conceptual high-level architecture documentation for the Projexa project.

Updated Aug 1, 2026
One-click install
npx skills add https://github.com/Samind-Samind/projexa --skill architecture-samind-samind
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: architecture
Source: https://github.com/Samind-Samind/projexa/tree/main/.claude/skills/architecture
Command: npx skills add https://github.com/Samind-Samind/projexa --skill architecture-samind-samind

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Designing a high-level system architecture requires structured decisions about components, boundaries, and data flows, and doing it ad hoc leads to inconsistent or technology-premature documentation. This Skill runs a disciplined, human-in-the-loop interview process to produce a conceptual architecture document that stays free of technology lock-in. ## Core Features & Use Cases - Structured architecture interview: Asks the user clarifying questions with at least three options (each with pros and cons) whenever information is not documented, never guessing on the user's behalf. - Conceptual deliverables: Produces a system context diagram, component/layer breakdown, mermaid data-flow diagrams derived from the real user journey, AI orchestration boundaries, and cross-cutting concerns (audit, RBAC, resilience). - Controlled file writing: Shows the full draft for review before writing, then delegates file creation to an architecture-writer sub agent that updates docs/02-design/02-technical/architecture.md and optionally syncs section 3 of the system design document. - Use Case: A PM or SA asks to draft the Projexa system architecture; the Skill reads the design document and user journey, interviews the user on undocumented decisions, presents diagrams for confirmation, and writes the confirmed architecture document. ## Quick Start Ask the assistant to create or update the high-level conceptual architecture document for the Projexa project.

Frequently Asked Questions about architecture

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

FAQPage Schema
How do I create a high-level architecture document for a project?▼

Run this Skill and specify a scope, such as the full document or a single data flow. It reads existing project documents, interviews you on undocumented decisions with at least three options each, shows a draft for review, and writes the confirmed result to the architecture document.

What does a conceptual architecture document include?▼

It includes a system context diagram, component and layer breakdown with responsibilities, mermaid data-flow diagrams derived from the user journey, AI orchestration boundaries, and cross-cutting concerns like audit trails, RBAC, and resilience, all without naming specific technologies.

Can the architecture document reference specific technologies or frameworks?▼

No. The document must stay conceptual and never name databases, languages, frameworks, or cloud providers. Technology selection is handled separately by the tech-stack skill, which should align with this document afterward.

What happens if project information is missing or undocumented?▼

The Skill never guesses. It asks the user via structured questions offering at least three approaches with pros and cons. If the user delegates the decision, the chosen option is labeled as AI-assessed on the user's behalf in the document.

When should I not use this architecture skill?▼

Do not use it for choosing real technologies, frameworks, or vendors; use the tech-stack skill instead. It also requires an existing user-journey document for data-flow sections, so run the journey generation skill first if that file is missing.