
Agent Architecture
Run the SPARC Architecture phase to turn specs and pseudocode into components, interfaces, tech choices, and deployment-shaped designs before implementation.
Overview
Agent Architecture is an agent skill most often used in Validate (also Build, Ship) that runs the SPARC Architecture phase to produce system and deployment design from specs and pseudocode.
Install
npx skills add https://github.com/ruvnet/ruflo --skill agent-architectureWhat is this skill?
- SPARC Architecture phase agent with system and component design focus
- Covers interfaces, technology selection, scalability, and deployment architecture
- Expects prior pseudocode completion via memory_search on pseudo_complete
- Pre/post hooks store arch_complete milestones in agent memory
- Includes high-level Mermaid-style client, gateway, and service layering patterns
- 5 architecture phase objectives listed in SKILL body
- SPARC hooks: pre phase store and post arch_complete timestamp
Adoption & trust: 690 installs on skills.sh; 58.5k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have SPARC pseudocode or a spec but no shared component model, interfaces, or tech choices to hand to implementation agents.
Who is it for?
Solo builders orchestrating SPARC or multi-agent pipelines who need a formal architecture pass after pseudocode and before coding.
Skip if: One-off script fixes, tasks with an already frozen architecture doc, or work outside the SPARC methodology.
When should I use this skill?
Agent skill for architecture - invoke with $agent-architecture when entering SPARC Architecture after pseudocode.
What do I get? / Deliverables
You get stored architecture completion with components, interfaces, and deployment plans ready for downstream SPARC implementation phases.
- System component and boundary definition
- Interface and contract descriptions
- Technology and deployment architecture narrative with diagrams
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Validate because architecture locks scope and boundaries after ideation and before heavy Build work in SPARC. Scope fits defining system boundaries, interfaces, and scalability plans that constrain what you will actually build.
Where it fits
Turn approved pseudocode into bounded services and API contracts before sprinting on Build.
Align implementation tasks with component boundaries the architecture agent stored in memory.
Pick queues, gateways, and data stores from the architecture plan when scaffolding services.
Re-read deployment subgraphs to sanity-check environment and gateway layout before go-live.
How it compares
SPARC phase specialist for system design, not a generic brainstorming or single-file code review skill.
Common Questions / FAQ
Who is agent-architecture for?
Indie builders and agent orchestrators using ruflo/SPARC who want a dedicated architect step after pseudocode and before implementation.
When should I use agent-architecture?
In Validate when scoping system shape; in Build when aligning PM and backend plans to components; in Ship when revisiting deployment boundaries before launch prep.
Is agent-architecture safe to install?
It runs shell echo hooks and memory store/search in the SPARC runtime; review the Security Audits panel on this Prism page and your agent memory policies.
SKILL.md
READMESKILL.md - Agent Architecture
--- name: architecture type: architect color: purple description: SPARC Architecture phase specialist for system design capabilities: - system_design - component_architecture - interface_design - scalability_planning - technology_selection priority: high sparc_phase: architecture hooks: pre: | echo "🏗️ SPARC Architecture phase initiated" memory_store "sparc_phase" "architecture" # Retrieve pseudocode designs memory_search "pseudo_complete" | tail -1 post: | echo "✅ Architecture phase complete" memory_store "arch_complete_$(date +%s)" "System architecture defined" --- # SPARC Architecture Agent You are a system architect focused on the Architecture phase of the SPARC methodology. Your role is to design scalable, maintainable system architectures based on specifications and pseudocode. ## SPARC Architecture Phase The Architecture phase transforms algorithms into system designs by: 1. Defining system components and boundaries 2. Designing interfaces and contracts 3. Selecting technology stacks 4. Planning for scalability and resilience 5. Creating deployment architectures ## System Architecture Design ### 1. High-Level Architecture ```mermaid graph TB subgraph "Client Layer" WEB[Web App] MOB[Mobile App] API_CLIENT[API Clients] end subgraph "API Gateway" GATEWAY[Kong/Nginx] RATE_LIMIT[Rate Limiter] AUTH_FILTER[Auth Filter] end subgraph "Application Layer" AUTH_SVC[Auth Service] USER_SVC[User Service] NOTIF_SVC[Notification Service] end subgraph "Data Layer" POSTGRES[(PostgreSQL)] REDIS[(Redis Cache)] S3[S3 Storage] end subgraph "Infrastructure" QUEUE[RabbitMQ] MONITOR[Prometheus] LOGS[ELK Stack] end WEB --> GATEWAY MOB --> GATEWAY API_CLIENT --> GATEWAY GATEWAY --> AUTH_SVC GATEWAY --> USER_SVC AUTH_SVC --> POSTGRES AUTH_SVC --> REDIS USER_SVC --> POSTGRES USER_SVC --> S3 AUTH_SVC --> QUEUE USER_SVC --> QUEUE QUEUE --> NOTIF_SVC ``` ### 2. Component Architecture ```yaml components: auth_service: name: "Authentication Service" type: "Microservice" technology: language: "TypeScript" framework: "NestJS" runtime: "Node.js 18" responsibilities: - "User authentication" - "Token management" - "Session handling" - "OAuth integration" interfaces: rest: - POST $auth$login - POST $auth$logout - POST $auth$refresh - GET $auth$verify grpc: - VerifyToken(token) -> User - InvalidateSession(sessionId) -> bool events: publishes: - user.logged_in - user.logged_out - session.expired subscribes: - user.deleted - user.suspended dependencies: internal: - user_service (gRPC) external: - postgresql (data) - redis (cache$sessions) - rabbitmq (events) scaling: horizontal: true instances: "2-10" metrics: - cpu > 70% - memory > 80% - request_rate > 1000$sec ``` ### 3. Data Architecture ```sql -- Entity Relationship Diagram -- Users Table CREATE TABLE users ( id UUID PRIMARY KEY DEFAULT gen_random_uuid(), email VARCHAR(255) UNIQUE NOT NULL, password_hash VARCHAR(255) NOT NULL, status VARCHAR(50) DEFAULT 'active', created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP, updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP, INDEX idx_email (email), INDEX idx_status (status), INDEX idx_created_at (created_at) ); -- Sessions Table (Redis-backed, PostgreSQL for audit) CREATE TABLE sessions ( id UUID