Microservices Architecture Planner skill for Claude and Codex
Designs a microservices architecture around your business domains: which services to cut, how they talk, what owns which data, and the infrastructure to run…
Install it once and Claude or Codex runs it for you. When your request matches, your AI loads the skill, asks for anything it is missing, and works through the template itself. Prefer to fill it in by hand? See the full prompt template.
Download the skill (.zip) · View SKILL.md
When it is used
Use when an app needs its business domains split into microservices with clear data ownership, communication patterns, and the infrastructure to run them. Do not use for designing a single service's internal code structure or writing Kubernetes manifests. Related topics: coding, architecture, microservices.
What it will ask you for
- Application type (free text, e.g. "E-commerce platform or SaaS dashboard")
- Current architecture (choose one: Monolith, Mini-services, Already microservices, Greenfield)
- Expected scale (choose one: Small (<1000 users), Medium (1000-100k users), Large (100k+ users), Massive (millions))
- Domains to separate into services (longer free text)
- Communication pattern (choose any: REST API, gRPC, GraphQL, Message queue (Kafka/RabbitMQ), Event sourcing)
- Container orchestration (choose one: Kubernetes, Docker Swarm, ECS (AWS), Cloud Run, None)
- Service mesh (choose one: Yes, No)
- API gateway (choose one: Yes, No)
- Include observability (logging and metrics and tracing) (choose one: Yes, No)
How it works
- Decide for each domain whether it earns a service by data model, change rate, or scaling profile, and merge the ones that do not.
- Assign each service the data it owns and the interfaces it exposes, so no two services write the same table.
- Choose sync or async communication per interaction and mark where idempotency and retries are required.
- Place the gateway, mesh, and observability stack where they take work off the services and name what each handles.
- Size the infrastructure for the expected scale, identify the first bottleneck, and name the first migration or build step.
What you get
A service table (name, responsibility, owned data, exposes, consumes), an interaction diagram in text or Mermaid, an infrastructure list sized to the scale, a first migration step, and a short risk list.
What it watches out for
- Cutting a service per database table or per team wish, which produces a distributed monolith that cannot deploy independently.
- Letting two services write the same data store, so consistency problems appear the first time they disagree.
- Choosing a mesh and gateway for a small-scale system where they add more operations load than they remove.
- Making every call synchronous, so one slow service drags down the whole user request.
How to install
Claude.ai and the Claude desktop app
- Download the skill .zip.
- Open Settings, then Capabilities, and find Skills.
- Choose Upload skill and select the .zip.
- Ask Claude for help with the task. It loads the skill when your request matches.
Claude Code
- Download the skill .zip.
- Unzip it into ~/.claude/skills/ (or .claude/skills/ inside a project).
- Start a new session. Claude Code loads the skill when your request matches.
Codex
- Download the skill .zip.
- Unzip it into ~/.agents/skills/ (or .agents/skills/ inside a project).
- Codex picks it up on its own. Type $ and the skill name to use it, or just ask for the task.
Cursor
- Download the Cursor rule (.mdc).
- Save it in .cursor/rules/ inside your project.
- Cursor's agent applies the rule when your request matches its description.
AGENTS.md (Codex, Cursor, and other coding agents)
- Copy the AGENTS.md section.
- Paste it into AGENTS.md at the root of your repository, creating the file if needed.
- Coding agents that read AGENTS.md follow it when your request fits.