Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsSoftware Architect Skill Pro
Architectural framework for system design, trade-off analysis, and technical decision-making.
$9.99
Software Architect Skill Pro
Example session with this skill installed
Propose an architecture for a real-time notification service that must handle 10k events/sec. We need high reliability and low latency. Provide trade-offs between Polling and WebSockets.
- Read your context and instructions
- Compiled the software architect pro
I have analyzed the requirements for the real-time notification service. Below is the architectural recommendation focusing on high-throughput event processing and the requested communication protocol trade-offs.
Executive Recommendation
For a 10k events/sec workload requiring low latency, I recommend an event-driven architecture using an asynchronous message broker (like Kafka or Pulsar) with a specialized WebSocket gateway layer for client delivery.
Architecture Overview
graph LR
P[Producers] --> B[Message Broker]
B --> C[Notification Service]
C --> D[(State Store)]
C --> G[WebSocket Gateway]
G --> U[End Users]
Options and Trade-off Table
| Feature | Option A: WebSockets | Option B: Long Polling |
|---|---|---|
| Latency | Real-time (low ms) | High (dependent on interval) |
| Connection Overhead | High initial, low per-message | High per-request |
| Server State | Stateful (requires affinity) | Stateless |
| Reliability | Requires heartbeat/reconnect logic | Naturally resilient to drops |
| Scalability | Complex (socket exhaustion limits) | Simple (horizontal scaling) |
Components and Responsibilities
- Message Broker: Decouples event ingestion from delivery, ensuring durability at 10k/sec.
- WebSocket Gateway: Manages persistent connections and pushes notifications to active clients.
- State Store: Tracks user preferences and delivery status for offline notifications.
Risks and Mitigations
Risk: Socket exhaustion on gateway nodes.
Mitigation: Implement horizontal scaling with a load balancer supporting sticky sessions.
Risk: Backpressure during traffic spikes.
Mitigation: Use consumer groups in the broker to scale notification workers dynamically.
Next steps
- Define the specific message schema for the API contract.
- Select a message broker based on existing infrastructure (e.g., AWS Kinesis vs. self-hosted Kafka).
- Conduct a load test to determine the maximum concurrent connections per gateway node.
Connects securely to your tools. The creator never sees your data.
About this skill
Building complex systems without a clear architectural roadmap leads to technical debt and expensive pivots. This skill provides a structured framework for planning, reviewing, and evolving software systems through explicit trade-off analysis. It transforms vague business goals into decision-ready technical specifications grounded in reality.
What it does
- Requirement extraction organizes functional needs, quality attributes, and constraints into a prioritized hierarchy.
- System mapping documents data flows, trust boundaries, and component ownership to visualize system surface area.
- Option analysis generates multiple viable architectures to compare different technical paths for the same problem.
- Trade-off evaluation assesses reliability, latency, cost, and maintainability against stated business priorities.
- Risk identification uncovers single points of failure, bottlenecks, and migration complexities before code is written.
How it works
- Frame the business goals, system boundaries, and success measures to align on the problem space.
- Propose 2 to 3 material architecture options, including a baseline and rejected alternatives.
- Compare choices using a structured trade-off table focusing on operational and delivery impacts.
- Detail the selected path with component responsibilities, API contracts, and failure handling strategies.
Frameworks & tools
This skill is framework-agnostic. It works across microservices, monoliths, serverless, and distributed systems. It outputs structured Markdown and Mermaid diagrams for architectural documentation.
Why this beats prompting it yourself
Generic prompts often overlook failure modes or ignore the operational burden of a design. This skill enforces a rigorous checklist including idempotency, backpressure, and reversibility that standard LLM responses frequently skip.
Use cases
- Designing a new distributed system from scratch with documented ADRs.
- Reviewing an existing architecture to identify scaling bottlenecks and security gaps.
- Planning a complex database or cloud migration with a staged rollout strategy.
- Comparing tech stacks for a new product launch based on team capacity and cost.
Known limitations
Does not perform live repository inspections or infrastructure deployments. Advice is for decision support and requires expert review for high-impact or regulated systems.
No live repository access or runtime validation. Numeric scoring is avoided unless criteria are user-defined. Requires expert review for security-critical systems.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 16 days ago
Permissions
No special permissions declared or detected