# Design Reliable Software Architectures

Complex systems require explicit boundaries, trade-offs, and operational constraints. This skill structures architecture analysis, diagrams, decisions, and documentation for review.

## Install

```bash
npx skillstore add zl2023github/software-architect
```

## Metadata

- Status: approved
- Slug: zl2023github-software-architect
- Skillstore revision: r2
- Version status: missing
- Tree hash: 1fe44cd2f6349c5ad8ec4a6fd71ec9c01f07292a105a3c18b5867b74f8ca3783
- Author: zl2023github
- GitHub username: zl2023github
- License: MIT
- Repository: https://github.com/zl2023github/software-engineer-skills/tree/main/software-architecture/software-architect
- Ref: 88a8e9a07f4c54ab105c1c41b6267c287146b07b
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 66
- Public page: https://skillstore.pages.dev/skills/zl2023github-software-architect
- Manifest: https://skillstore.pages.dev/api/skills/zl2023github-software-architect/manifest

## Capabilities

- Designs system boundaries, modules, services, data flows, deployment structures, and security controls.
- Creates C4, UML, deployment, data-flow, and entity-relationship diagram specifications.
- Writes architecture decision records with context, decisions, consequences, and trade-offs.
- Compares technologies across functionality, performance, cost, community support, learning requirements, and risk.
- Analyzes codebase dependencies, architecture styles, technical debt, and architecture compliance.
- Produces structured architecture, API, deployment, and technical proposal documents.

## Use Cases

- Design a New Platform: Turn product requirements into service boundaries, data flows, deployment choices, diagrams, and documented trade-offs.
- Evaluate a Technology Decision: Compare candidate technologies using explicit criteria, risks, proof-of-concept steps, and a recommended decision.
- Review an Existing System: Assess dependencies, architecture style, technical debt, compliance gaps, and practical modernization priorities.

## Prompt Templates

### Outline a System Architecture

```
Design an architecture for [product]. Ask for missing requirements, then define components, data stores, integrations, deployment, security, and key trade-offs.
```

### Compare Technology Options

```
Compare [option A] and [option B] for [workload]. Evaluate performance, reliability, cost, operations, team fit, migration risk, and proof-of-concept criteria.
```

### Review a Codebase Architecture

```
Analyze the provided repository for module boundaries, dependency direction, coupling, data ownership, deployment structure, and technical debt. Prioritize findings by impact.
```

### Plan a System Modernization

```
Create an incremental modernization plan for [current system] under [constraints]. Include target architecture, transition stages, compatibility controls, rollback points, metrics, and ADRs.
```

## Limitations

- Recommendations do not replace performance tests, security reviews, compliance checks, or production validation.
- Results depend on accurate requirements, constraints, traffic estimates, team skills, and codebase context.
- External research, command execution, and file changes depend on available tools and user authorization.
- Mermaid and PlantUML output may require a compatible renderer and manual layout review.

## Best Practices

- Provide business goals, constraints, expected scale, reliability targets, compliance needs, and team capabilities.
- Request alternatives with explicit trade-offs, decision criteria, assumptions, and validation steps.
- Review generated diagrams and decisions with engineering, security, operations, data, and product stakeholders.

## Anti Patterns

- Do not select microservices before validating domain boundaries, scale needs, and operational capacity.
- Do not accept diagrams without ownership, failure behavior, data flows, and deployment assumptions.
- Do not treat generated recommendations as proof of performance, security, compliance, or production readiness.

## Security Audit

- Audited at: 2026-07-24T00:09:20.181\+00:00
- Summary: All 18 static findings are false positives caused by Markdown backticks, Mermaid diagrams, and high entropy in Chinese text. A separate medium-risk issue remains because the skill requests broad terminal, web, file-write, code-execution, and delegation access without approval boundaries. No prompt injection, obfuscation, or malicious commands were found.

## Stats

- Views: 3
- Downloads: 9
- Favorites: 0
- Popularity score: 0