T11 Software Architect
Paste the following prompt into your AI chat to install this skill:
Please install @user_b84755b2/sunnyxxx according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem This Skill Targets
Architecture discussions often stall between guesswork-based selection and over-engineering: teams may jump into microservices without clear capacity, fault domains, cost, compliance, or evolution constraints, or they reduce a platform-level design to a single caching question. This skill turns architecture consulting into an executable workflow so engineers can choose the right depth for point issues, surface issues, and asset issues, avoiding mismatched scope.
How It Works
It assumes mainstream public cloud as the default deployment context and follows a hard selection order: cloud-native managed services first, mainstream open-source HA deployments second, and private cloud/IDC options only when compliance, cost, or performance justify them. Key steps include:
- Scope classification: decide whether the task is a single optimization, a system design, or a reusable productized platform.
- First-principles analysis: examine business essence, data essence, physical limits, failure points, and 3-year evolution, with bounded depth.
- Complexity and capacity: identify dominant complexity and estimate QPS, storage, cost, and extension paths.
- Research and alternatives: compare public documentation, engineering blogs, and similar systems, then produce 3 to 5 meaningfully different designs.
- Selection and reliability: rank quality attributes, then define SLO, rate limiting, circuit breaking, degradation, canary release, rollback, and alerting.
- Delivery review: externalize unresolved assumptions and deliver Markdown, ADRs, Mermaid source files, and SVG diagrams.
Limits and Caveats
It is best for reviewable, traceable technical proposals rather than ad-hoc answers or implementation-only work. For highly custom business domains, sensitive compliance requirements, or unclear traffic volumes, key assumptions, SLOs, and cost boundaries must still be confirmed by stakeholders; it provides an architecture decision framework, not a substitute for official product docs or production load tests.
Use Cases
- Before launching a new order system, draft a reviewable plan covering capacity, selection, HA, and evolution.
- For slow order-query optimization, decide quickly whether caching, indexing, or sharding is necessary.
- When reviewing a microservice design, add SLO, throttling, circuit breaking, degradation, and open assumptions.
- When designing an AI Agent or RAG rollout, choose workflows, components, and cloud-native managed priorities.
Best For
- Backend architects owning core transaction systems who need pre-launch selection, capacity, and HA proposals.
- R&D leads running design reviews who need to catch gaps and externalize open assumptions.
- Solutions engineers handling customer scale questions who need to convert scenarios into component and cost estimates.
- Technical leads preparing replatforming who need two-phase evolution and migration paths.
Related Skills
Automatically indexes Gradle-cached AAR/JAR dependency classes and returns library coordinates, versions, and public APIs by fully qualified name, using only the Python standard library.
Codifies AMT and YourMT3 training conventions, script patterns, hyperparameters, precision, checkpoints, and NaN safeguards.
Retrieve relevant chunks from a customer-managed PKM dataset by dataset_id and return concise, source-annotated answers.
Convert PRDs, user stories, or functional specs into prioritized test-point checklists covering functional, business-rule, boundary, exception, and non-functional dimensions.