[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-80006":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":15,"stars30d":16,"stars90d":14,"forks30d":14,"starsTrendScore":17,"compositeScore":18,"rankGlobal":9,"rankLanguage":9,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":22,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":32,"readmeContent":33,"aiSummary":34,"trendingCount":14,"starSnapshotCount":14,"syncStatus":17,"lastSyncTime":35,"discoverSource":36},80006,"aws-architecture-diagram-skill","vidanov\u002Faws-architecture-diagram-skill","vidanov","Reusable skill for generating AWS architecture diagrams in draw.io format. Works with Kiro CLI and Claude Code.",null,"Python",90,7,1,0,4,18,2,46.51,"MIT License",false,"main",true,[24,25,26,27,28,29,30,31],"ai-tools","architecture-diagram","aws","claude-code","drawio","infrastructure-as-code","kiro-cli","serverless","2026-06-12 04:01:26","# AWS Architecture Diagram Skill\n\n[![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-vidanov.github.io-8C4FFF)](https:\u002F\u002Fvidanov.github.io\u002Faws-architecture-diagram-skill\u002F)\n\n![Prompt → AI Agent + Skill → .drawio diagram](docs\u002Fhero.png)\n\nA reusable skill for generating AWS architecture diagrams in draw.io format. Works with **any AI assistant** — ChatGPT, Claude Projects, Kiro CLI, Claude Code, or any internal AI agent that supports custom prompts and file uploads.\n\nGenerates production-quality `.drawio` files using official AWS Architecture Icons with proper styling, layout, and color coding.\n\n## Features\n\n- **Left-to-right flow** — UI\u002FFrontend on left, data sources on right\n- **Official AWS icons** — from draw.io's built-in `mxgraph.aws4` stencil library (sourced from [AWS Architecture Icons](https:\u002F\u002Faws.amazon.com\u002Farchitecture\u002Ficons\u002F))\n- **Verified icon catalog** — only icons confirmed to render correctly\n- **Consistent styling** — 78px icons, strokeWidth=2 edges, proper AWS color palette\n- **Export support** — PNG\u002FSVG\u002FPDF via draw.io Desktop CLI with embedded XML\n\n## Installation\n\n### Skills CLI (recommended — works with Claude Code, Cursor, Codex, Kiro, and 50+ agents)\n\n```bash\nnpx skills add vidanov\u002Faws-architecture-diagram-skill\n```\n\nInstall to a specific agent:\n\n```bash\nnpx skills add vidanov\u002Faws-architecture-diagram-skill --agent claude-code\nnpx skills add vidanov\u002Faws-architecture-diagram-skill --agent cursor\nnpx skills add vidanov\u002Faws-architecture-diagram-skill --agent kiro\n```\n\nInstall globally (available in all projects):\n\n```bash\nnpx skills add vidanov\u002Faws-architecture-diagram-skill --global\n```\n\n[![skills.sh](https:\u002F\u002Fskills.sh\u002Fb\u002Fvidanov\u002Faws-architecture-diagram-skill)](https:\u002F\u002Fskills.sh\u002Fvidanov\u002Faws-architecture-diagram-skill)\n\n### Any AI Assistant (ChatGPT, Claude Projects, Internal Agents)\n\nNo code or CLI required. Works with any AI that supports custom instructions and file uploads:\n\n1. **Copy the prompt** from [`chatgpt\u002FPROMPT.md`](chatgpt\u002FPROMPT.md) into your assistant's system prompt \u002F custom instructions field\n2. **Upload the reference files** from [`chatgpt\u002Freferences\u002F`](chatgpt\u002Freferences\u002F) to the knowledge base \u002F project files\n3. **Start asking** for AWS architecture diagrams\n\nThis works with:\n- **ChatGPT** → Custom GPT or Project with instructions + files\n- **Claude Projects** → Project instructions + project knowledge\n- **Internal corporate AI agents** → System prompt + knowledge base uploads\n- **Any LLM with file context** → Paste prompt + reference content\n\n> **Tip:** Set temperature to 0.3 for consistent XML output.\n\n### Claude Code (Plugin — recommended)\n\n```bash\n\u002Fplugin marketplace add vidanov\u002Faws-architecture-diagram-skill\n\u002Fplugin install aws-architecture-diagram@vidanov-skills\n```\n\nThen use it:\n```\n\u002Faws-architecture-diagram:aws-architecture-diagram\n```\n\nOr just ask naturally — the skill activates automatically when you mention AWS architecture diagrams.\n\n### Claude Code (Manual)\n\n```bash\nmkdir -p ~\u002F.claude\u002Fskills\u002Faws-architecture-diagram\ncp claude\u002FSKILL.md ~\u002F.claude\u002Fskills\u002Faws-architecture-diagram\u002FSKILL.md\ncp -r references ~\u002F.claude\u002Fskills\u002Faws-architecture-diagram\u002Freferences\n```\n\n### Kiro CLI\n\n```bash\n# Global (all projects)\nmkdir -p ~\u002F.kiro\u002Fskills\u002Faws-architecture-diagram\ncp kiro\u002FSKILL.md ~\u002F.kiro\u002Fskills\u002Faws-architecture-diagram\u002FSKILL.md\ncp -r references ~\u002F.kiro\u002Fskills\u002Faws-architecture-diagram\u002Freferences\n\n# Per-project\nmkdir -p .kiro\u002Fskills\u002Faws-architecture-diagram\ncp kiro\u002FSKILL.md .kiro\u002Fskills\u002Faws-architecture-diagram\u002FSKILL.md\ncp -r references .kiro\u002Fskills\u002Faws-architecture-diagram\u002Freferences\n```\n\n## Usage\n\nJust ask to create an AWS architecture diagram:\n\n```\nCreate an AWS architecture diagram for a serverless API with Lambda, DynamoDB, and API Gateway\n```\n\nOr with export:\n\n```\nCreate an AWS architecture diagram as PNG for a real-time data pipeline with Kinesis, Lambda, and S3\n```\n\n## Example Output\n\n### Event-Driven Order Processing\n> \"Create an event-driven order processing architecture with SQS, Lambda, DynamoDB, and EventBridge\"\n\n![Event-Driven Order Processing](docs\u002Fexample-event-driven.png)\n[Download .drawio](examples\u002Fexample-event-driven.drawio)\n\n### Real-Time IoT Analytics\n> \"Create a real-time IoT analytics pipeline with Kinesis, Lambda, S3 data lake, and DynamoDB\"\n\n![Real-Time IoT Analytics](docs\u002Fexample-streaming.png)\n[Download .drawio](examples\u002Fexample-streaming.drawio)\n\n### 3-Tier Web Application\n> \"Create a 3-tier web application with CloudFront, ALB, ECS Fargate, Aurora, and ElastiCache\"\n\n![3-Tier Web Application](docs\u002Fexample-3tier.png)\n[Download .drawio](examples\u002Fexample-3tier.drawio)\n\n## Structure\n\n```\naws-architecture-diagram-skill\u002F\n├── README.md\n├── LICENSE\n├── chatgpt\u002F\n│   ├── PROMPT.md                       # Universal prompt (ChatGPT, Claude Projects, any AI)\n│   └── references\u002F                     # Same icons in .txt format for upload\n│       ├── aws-icons-common.txt\n│       ├── aws-icons-compute.txt\n│       ├── aws-icons-database.txt\n│       ├── aws-icons-integration.txt\n│       ├── aws-icons-networking.txt\n│       ├── aws-icons-storage.txt\n│       ├── aws-icons-security.txt\n│       ├── aws-icons-analytics-ml.txt\n│       └── aws-icons-iot-migration-devtools.txt\n├── kiro\u002F\n│   └── SKILL.md                        # Kiro CLI version\n├── claude\u002F\n│   └── SKILL.md                        # Claude Code version\n└── references\u002F\n    ├── aws-icons-compute.md            # Lambda, EC2, ECS, EKS, Fargate (25+ icons)\n    ├── aws-icons-database.md           # DynamoDB, RDS, Aurora, ElastiCache (40+ icons)\n    ├── aws-icons-integration.md        # API GW, SQS, SNS, EventBridge, Step Functions (35+ icons)\n    ├── aws-icons-networking.md         # CloudFront, Route 53, VPC, ELB (40+ icons)\n    ├── aws-icons-storage.md            # S3, EFS, EBS, Glacier, Backup (40+ icons)\n    ├── aws-icons-security.md           # IAM, Cognito, KMS, WAF, Shield (45+ icons)\n    ├── aws-icons-analytics-ml.md       # Kinesis, Athena, Bedrock, SageMaker (45+ icons)\n    └── aws-icons-common.md             # Groups, general resources, edge styles, base template\n```\n\n## Key Insight: Two Icon Patterns\n\ndraw.io AWS icons have **two patterns** with opposite `strokeColor` rules:\n\n| Pattern | Style | strokeColor | Use for |\n|---------|-------|-------------|---------|\n| Service-level | `shape=mxgraph.aws4.resourceIcon;resIcon=mxgraph.aws4.\u003Cname>` | `#ffffff` (REQUIRED) | Main service icons (colored square + white glyph) |\n| Resource-level | `shape=mxgraph.aws4.\u003Cname>` | `none` (REQUIRED) | Sub-resources (silhouette icons) |\n\n**Confusing these patterns is the #1 cause of broken icons in AI-generated diagrams.**\n\n## Key Insight: Legacy Stencil Names\n\ndraw.io stencil names don't always match current AWS service names. Renamed services keep their old stencil identifiers:\n\n| AWS Service | draw.io `resIcon` | Reason |\n|---|---|---|\n| Amazon OpenSearch Service | `elasticsearch_service` | Renamed from Elasticsearch (2021) |\n\n**Empty icon = wrong stencil name.** The skill includes a verified icon catalog to prevent this.\n\n## Two-Step Edit Approach\n\nThe skill instructs the AI agent to:\n1. Generate the `.drawio` XML\n2. Export to PNG and visually review for broken icons or layout issues\n3. Fix and re-export\n\nThis catches rendering problems that are invisible in raw XML.\n\n## Supported Services (110+ icons)\n\n| Category | Icons |\n|----------|-------|\n| Compute | Lambda, EC2, ECS, EKS, Fargate |\n| App Integration | API Gateway, SNS, SQS, EventBridge, Step Functions |\n| Database | DynamoDB, RDS, Aurora, ElastiCache |\n| Storage | S3, EFS, EBS |\n| Networking | CloudFront, Route 53, VPC, ELB (ALB\u002FNLB) |\n| Security | IAM, Cognito, KMS, WAF |\n\n## Known Broken Icons\n\nThese `resIcon` values do NOT render in draw.io — avoid them:\n\n- `mxgraph.aws4.dynamodb_table` → use `mxgraph.aws4.dynamodb` instead\n- `mxgraph.aws4.dynamodb_stream` → use `mxgraph.aws4.dynamodb` with label\n- `mxgraph.aws4.general_saml_token` → use `mxgraph.aws4.traditional_server`\n\n---\n\n## Comparison with Alternatives\n\n| Solution | Type | Output | Editable | Zero-deps | Verified Icons |\n|----------|------|--------|----------|-----------|----------------|\n| **[aws-architecture-diagram-skill](https:\u002F\u002Fgithub.com\u002Fvidanov\u002Faws-architecture-diagram-skill)** | Agent Skill | `.drawio` | ✅ | ✅ | ✅ |\n| [jgraph\u002Fdrawio-mcp](https:\u002F\u002Fgithub.com\u002Fjgraph\u002Fdrawio-mcp) | MCP Server + Skill | `.drawio` | ✅ | ❌ (MCP) | ❌ |\n| [awslabs\u002Faws-diagram-mcp-server](https:\u002F\u002Fpypi.org\u002Fproject\u002Fawslabs.aws-diagram-mcp-server\u002F) | MCP Server (Python) | PNG | ❌ | ❌ (Python+GraphViz) | N\u002FA |\n| [awslabs\u002Fdiagram-as-code](https:\u002F\u002Fgithub.com\u002Fawslabs\u002Fdiagram-as-code) | CLI (Go) | PNG\u002FSVG | ❌ | ❌ (Go binary) | N\u002FA |\n| [carlosmgv02\u002Fdiagram-ai-generator](https:\u002F\u002Fgithub.com\u002Fcarlosmgv02\u002Fdiagram-ai-generator) | MCP Server | Multi-cloud | ❌ | ❌ (MCP) | ❌ |\n| [clouda.ai](https:\u002F\u002Fclouda.ai\u002F) | SaaS | PNG | ❌ | N\u002FA | N\u002FA |\n\n**Why this approach wins:**\n- No runtime dependencies. It's a markdown file, not a server.\n- Output is native `.drawio` XML you can open, edit, and version-control.\n- Verified icon catalog documents which `mxgraph.aws4.*` names actually render vs. silently break.\n- Opinionated layout: left-to-right flow with consistent spacing, not random placement.\n- Works with any AI assistant: ChatGPT, Claude, Kiro CLI, or internal corporate agents.\n\n## Contributing\n\nContributions are welcome! Here's how you can help:\n\n1. **Fork** this repository\n2. **Create a branch** for your feature or fix (`git checkout -b my-feature`)\n3. **Make your changes** — add icons, fix stencil names, improve layout rules, etc.\n4. **Submit a Pull Request** with a clear description of what you changed and why\n\nYou can also **open an Issue** if you:\n- Found a broken icon or wrong stencil name\n- Want to request support for a new AWS service\n- Have suggestions for layout or styling improvements\n- Encountered a bug with a specific AI agent\n\n## License\n\nMIT\n","该项目是一个可重用的技能，用于生成draw.io格式的AWS架构图。它支持与Kiro CLI和Claude Code等工具集成，并且可以与任何支持自定义提示和文件上传的人工智能助手一起工作。其核心功能包括使用官方AWS图标库创建具有正确样式、布局和颜色编码的专业级.drawio文件，支持从左至右的数据流表示，以及通过draw.io桌面CLI导出为PNG\u002FSVG\u002FPDF格式。适用于需要快速可视化AWS基础设施设计的场景，如项目规划、技术交流或文档编制，特别适合于那些希望利用AI辅助提高工作效率的技术团队。","2026-06-11 03:58:53","CREATED_QUERY"]