Architecture Diagram logo

Architecture Diagram

Organization
proyecto26
architecture-diagram

This skill should be used when a system design needs a diagram — "draw the architecture", "diagram this system", "show the components", "make an architecture/infrastructure/topology diagram", or visualizing boxes-and-arrows, data flow, regions, or failure paths for a design. It generates a self-contained dark-theme HTML + SVG diagram (with PNG/PDF export). Use it whenever the `system-design` orchestrator or a building block reaches the "draw it" step, even if the user doesn't say "diagram".

Overview

Publisherproyecto26
Repositorysystem-design-skills
Skill namearchitecture-diagram
Stars
117
Forks
14
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by proyecto26 on GitHub. Read the source before you install it.

Installation

Install the Architecture Diagram AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/proyecto26/system-design-skills/tree/main/skills/architecture-diagram
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/proyecto26/system-design-skills.git /tmp/system-design-skills
mkdir -p .claude/skills
cp -r /tmp/system-design-skills/skills/architecture-diagram .claude/skills/architecture-diagram
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Architecture Diagram in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Architecture Diagram on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Architecture Diagram is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Architecture Diagram

Turn a stabilized design — a list of components and the connections between them — into a polished, self-contained HTML + SVG diagram. Components are boxes, arrows are communication/data flow, dashed boundaries are regions or trust zones. This is the diagramming engine for the whole plugin; the orchestrator and building blocks hand off here when it's time to draw.

Self-contained by design: one HTML file with inline SVG and CSS, a system monospace font (no web-font fetch), and no dependency to render — it draws fully offline. The only external calls are two pinned, SRI-protected CDN scripts that power the optional PNG/PDF export; remove them and the diagram still renders — only the export buttons are lost.

When to reach for this

At step 3 (high-level design) and step 6 (deep-dive) of the reasoning loop, and any time a design is clear enough to draw: the component list and their connections are known. Also for failure-path diagrams (step 5) — showing a degradation flow is part of the design.

When NOT to

Before the design has stabilized — don't draw boxes that aren't yet justified (that's failure mode #7). A quick ASCII sketch in the conversation is fine for early thinking; render with this skill once the components have earned their place.

How to build a diagram

  1. Gather the spec from the design: the list of components (each with a type), the directed connections (with protocol/label), and any region/boundary groupings. The system-design design-doc already produces this.
  2. Copy the template assets/template.html to a working file. It carries the dark theme, grid background, fonts, the export toolbar, and the summary-card layout — keep those intact.
  3. Map each component to a semantic type (color) using the table below.
  4. Place boxes and draw arrows following the layout/spacing rules in references/design-system.md (draw arrows before boxes so they sit behind; mask arrows behind semi-transparent fills; keep ≥40px vertical gaps; legends go outside boundary boxes).
  5. Label the flow — number arrows to show request order; use dashed arrows for async/replication/fallback; dashed boundaries for regions and security groups.
  6. Fill the summary cards with the key decisions/trade-offs (ties to the design doc), update the title/footer.
  7. Open it: open <file>.html to view; the toolbar exports PNG/PDF.

Semantic colors (map components to types)

These align with how the building blocks think about a system.

Component typeFill (rgba)StrokeUsed for
Frontend / clientrgba(8,51,68,0.4)#22d3eeweb/mobile clients, edge
Backend / servicergba(6,78,59,0.4)#34d399app/API services, workers
Database / storergba(76,29,149,0.4)#a78bfaSQL/NoSQL, object store
Cachergba(8,51,68,0.4)#38bdf8Redis/Memcached, CDN cache
Message bus / queuergba(251,146,60,0.3)#fb923cKafka, SQS, queues, streams
Cloud / managedrgba(120,53,15,0.3)#fbbf24managed services, regions
Security / authrgba(136,19,55,0.4)#fb7185gateways, auth, firewalls
External / genericrgba(30,41,59,0.5)#94a3b8third parties, DNS

Full styling, arrow markers, masking, spacing, and legend rules are in references/design-system.md.

Output rules

  • One self-contained .html file: inline CSS, inline SVG, no external images.
  • Keep the two pinned CDN scripts (html2canvas + jsPDF) with their SRI hashes — they power Copy/PNG/PDF export and nothing else. The diagram renders fully without them; they're only needed for export.
  • Avoid SVG <foreignObject> (renders inconsistently in html2canvas) — use plain <rect>/<text>.
  • Keep it to ~8–15 components for a high-level view; zoom into one subsystem for a deep-dive rather than cramming everything into one picture.

Dos and don'ts

Do:

  • Draw only after the design stabilizes; render the component list and connections the design doc already produced.
  • Map every box to a semantic type from the color table so the diagram reads at a glance.
  • Draw arrows before boxes (so they sit behind), number them for request order, and reserve dashed arrows for async/replication/fallback.
  • Keep it to ~8–15 components; zoom into one subsystem for a deep-dive instead of one crowded picture.
  • Keep the file self-contained — inline CSS/SVG, system monospace font, no external images — so it renders offline.

Don't:

  • Don't embed Mermaid or invent a second diagram convention; this skill is the one engine.
  • Don't use SVG <foreignObject> (html2canvas renders it inconsistently) — use plain <rect>/<text>.
  • Don't strip or rewrite the pinned CDN scripts' SRI hashes; recompute them only when bumping versions.
  • Don't place legends inside boundary boxes or stack components with under-40px gaps — both cause overlaps.
  • Don't add a hard dependency on any send-to-Claude transport; the Copy-Prompt button is the self-contained default.

Interactive / collaborative mode (optional)

For iterating with the user, an interactive variant adds layer toggles, connection-type filters, zoom, and click-to-comment that builds a feedback prompt with a Copy button. See references/interactive.md. It degrades gracefully: the Copy-Prompt button always works. A live "Send to Claude" button is possible only if a separate prompt-transport plugin (e.g. playground-sync) is installed — this plugin does not require or bundle one.

Related building blocks

  • system-design — invoked by the orchestrator at the "draw it" steps; its diagramming notes feed what to draw at each step.
  • Any building block — invoked by a block whose "Diagram" section says "render with architecture-diagram". This skill renders for all of them; it owns diagram generation and they point here rather than ship their own.

References

  • references/design-system.md — colors, arrow markers, z-order/masking, spacing, legend placement, the export toolbar internals. Read before drawing.
  • references/interactive.md — the optional interactive/commentable diagram pattern and the (optional) send-to-Claude transport. Read only if building an interactive map.

Assets

  • assets/template.html — the self-contained dark-theme HTML+SVG starting point with the export toolbar. Copy and customize it.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Architecture Diagram AI skill do?

This skill should be used when a system design needs a diagram — "draw the architecture", "diagram this system", "show the components", "make an architecture/infrastructure/topology diagram", or visualizing boxes-and-arrows, data flow, regions, or failure paths for a design. It generates a self-contained dark-theme HTML + SVG diagram (with PNG/PDF export). Use it whenever the `system-design` orchestrator or a building block reaches the "draw it" step, even if the user doesn't say "diagram".

Why use Architecture Diagram on TypingMind?

Because you install it once and use it with any model. Architecture Diagram is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Architecture Diagram in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/proyecto26/system-design-skills/tree/main/skills/architecture-diagram. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Architecture Diagram?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Architecture Diagram?

As many as you like. As long as a model supports skills, you can use Architecture Diagram with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Architecture Diagram AI skill free?

Yes. It is published on GitHub by proyecto26 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇