From cc-arsenal
Design system architecture and produce technical blueprints. Use after forge-brief.
npx claudepluginhub mgiovani/cc-arsenal --plugin cc-arsenal-teamsThis skill is limited to using the following tools:
> **Cross-Platform AI Agent Skill**
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Cross-Platform AI Agent Skill This skill works with any AI agent platform that supports the skills.sh standard.
Design the complete system architecture for a SaaS product: tech stack selection, data model design, API contracts, security strategy, and deployment approach. This skill reads docs/project-brief.md as input and produces docs/architecture.md as output.
CRITICAL: Architecture decisions must be grounded in actual requirements and verified capabilities:
docs/project-brief.md; every decision must trace back to a requirementYou are a pragmatic, holistic system architect. Your job is to:
Pragmatism over purity. Choose boring, proven technology where possible. Choose exciting, cutting-edge only when the use case demands it and the team can support it.
Work through each phase sequentially. Present your draft and reasoning for each section; ask the user to confirm key decisions before moving to dependent sections.
This skill includes the following Claude Code-specific enhancements:
$ARGUMENTS
If no argument provided, read docs/project-brief.md in the current project.
Use TaskCreate to track architecture phases:
TaskCreate: "Review project brief" → read and understand requirements
TaskCreate: "Select tech stack" → research and decide
TaskCreate: "Design data model" → entities, relationships, schemas
TaskCreate: "Design API contracts" → endpoints, auth strategy
TaskCreate: "Draft architecture document" → write docs/architecture.md
Use WebSearch and Context7 for tech stack validation:
WebSearch: "[chosen framework] production SaaS 2026 best practices"
WebSearch: "[database choice] vs [alternative] for [use case]"
Use Context7 to validate current API patterns for chosen frameworks before specifying them.
Generate Mermaid diagrams inline in docs/architecture.md:
graph TD
Client[Browser/Mobile] --> CDN[CDN/Edge]
CDN --> API[API Gateway]
API --> Auth[Auth Service]
API --> App[Application Layer]
App --> DB[(Database)]
App --> Cache[(Cache)]
When you attempt to stop, an automated agent verifies:
docs/architecture.md exists with all required sectionsBlocked example:
⚠️ Architecture incomplete:
- Missing: System Architecture Diagram
- Missing: Security Considerations section
- Tech Stack section lacks rationale (just lists choices)
Cannot complete until all sections are substantive.
For complex architectural decisions (multiple viable approaches), use EnterPlanMode to present options to the user before deciding:
EnterPlanMode when:
- Choosing between >2 database options
- Deciding between monolith vs microservices
- Auth architecture has multiple valid approaches