Draw.io AWS
AWS architecture diagrams from a real stencil catalog, not guessed icon names
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 885c256
Task: serverless API diagram — CloudFront → API Gateway → Lambda in private VPC subnet → RDS, plus S3 for static assets. Baseline (hand-written .drawio) produced 4 out of 5 incorrect colors against the catalog — CloudFront #F58534 instead of true #8C4FFF, API Gateway #D45B07 instead of #E7157B, Lambda #F58534 instead of #ED7100, RDS #2E73B8 instead of #C925D1 — used incorrect idiom `shape=mxgraph.aws4.<name>` instead of `shape=aws4.resource;resIcon=...`, lost Region/AZ containers, and placed S3 inside the private subnet, although S3 is not a subnet resident. Build via skill engine: validate ok, resIcons 5, grIcons 5, advice empty — against baseline resIcons 0, grIcons 0 and one advice. Main honesty caveat: `drawio-ai validate` returned ok:true even for the broken baseline — it doesn't catch stencil colors and idiom, the only clue is resIcons:0, so the claimed color validation in the description oversells. Render/vision self-check could not be run: it requires the desktop application draw.io ('draw.io CLI not found'), the step is unavailable in a headless environment — PNG not checked. Another defect: SKILL.md instructs to read $ROOT/docs/api-cheatsheet.md, but `docs` is not included in files[] package.json, and after the recommended `npm i -g`, this file is not on disk.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 3/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 3/5
What Draw.io AWS does
Builds AWS architecture .drawio diagrams — VPC/networking, event-driven, landing zone, multi-AZ, serverless — via the drawio-ai CLI, using ground-truth mxgraph.aws4 stencils and official category colors instead of invented shape names. Triggers when you ask for an AWS diagram. Requires a separate global npm install of the kit, and the draw.io desktop app for the PNG render step.
How to install Draw.io AWS
git clone https://github.com/sparklabx/drawio-ai-kit.git
mkdir -p ~/.claude/skills
cp -r drawio-ai-kit/skills/drawio-aws ~/.claude/skills/drawio-aws
npm i -g github:sparklabx/drawio-ai-kit
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Draw.io AWS
-
/drawio-awsAWS architecture diagrams from a real stencil catalog, not guessed icon names
It also activates on plain-language prompts like these:
-
Draw a serverless AWS architecture diagram with correct service icon colors -
Build a VPC networking diagram using real AWS stencils, not guessed shapes -
Create a multi-AZ landing zone diagram I can open in draw.io
Frequently asked questions
- Is the Draw.io AWS skill free?
- Yes. The skill itself is free from sparklabx/drawio-ai-kit. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Draw.io AWS work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Works with setup. Task: serverless API diagram — CloudFront → API Gateway → Lambda in private VPC subnet → RDS, plus S3 for static assets. Baseline (hand-written .drawio) produced 4 out of 5 incorrect colors against the catalog — CloudFront #F58534 instead of true #8C4FFF, API Gateway #D45B07 instead of #E7157B, Lambda #F58534 instead of #ED7100, RDS #2E73B8 instead of #C925D1 — used incorrect idiom `shape=mxgraph.aws4.<name>` instead of `shape=aws4.resource;resIcon=...`, lost Region/AZ containers, and placed S3 inside the private subnet, although S3 is not a subnet resident. Build via skill engine: validate ok, resIcons 5, grIcons 5, advice empty — against baseline resIcons 0, grIcons 0 and one advice. Main honesty caveat: `drawio-ai validate` returned ok:true even for the broken baseline — it doesn't catch stencil colors and idiom, the only clue is resIcons:0, so the claimed color validation in the description oversells. Render/vision self-check could not be run: it requires the desktop application draw.io ('draw.io CLI not found'), the step is unavailable in a headless environment — PNG not checked. Another defect: SKILL.md instructs to read $ROOT/docs/api-cheatsheet.md, but `docs` is not included in files[] package.json, and after the recommended `npm i -g`, this file is not on disk.
- What is the Draw.io AWS SkillProof Score?
- 7.6/10 — installs cleanly 3/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 3/5.
- How do I install Draw.io AWS?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Draw.io AWS with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.