name: architecture-diagram version: 1.5.0 description: "Build interactive, click-through architecture diagrams for software systems — a self-contained single HTML file with animated step-by-step flows, mode toggles (dev/prod, offline/online, v1/v2), dark/light theme, and a side panel with payload details, plus a companion markdown description. Use when the user wants to visualize or design a system: architecture diagram, service map, data flow, RAG/agentic flow, microservices topology, integration diagram, CI/CD or data/ETL pipeline, multi-agent system, or onboarding diagram — or to design a new service's flows before building. Natural-language triggers: 'I'm building X, what should the flow look like', 'pokaż jak działa system', 'diagram architektury', 'wizualizacja systemu', 'klikany diagram'. Do NOT use for static diagrams that belong inline (Mermaid/PlantUML), slide decks, or printable/PDF handouts — this produces interactive HTML for browser consumption." license: MIT

Interactive Architecture Diagrams

Build single-file, drop-in HTML pages that let workshop attendees, clients, or new team members click through how a system works — step by step, with animated data packets flowing between nodes, payload details on a side panel, and toggleable modes (offline/online, dev/prod, v1/v2). The aesthetic is dark and didactic-first: bounded nodes, gentle quadratic wires, packets that glow only on the active step — no rainbow gradients. Rebrand via the CSS tokens in assets/css-tokens.css.

When to reach for this skill

The differentiator is interactivity + sequenced data flow: if the value is "click through it and watch what happens", use this skill; if you need a static image, a slide, or an inline diagram, use Mermaid instead.

The mental model

Every diagram has four things:

1. Nodes — services, datastores, users, queues, external systems. Each has a role (color) and metadata (tech stack, port, deployment target).
2. Flows — named scenarios the user can pick (e.g. RAG Query, Direct Query, Ingest, Auth). Each flow is an ordered list of steps.
3. Steps — {from, to, color, title, route, payload, desc, chips}. Each step lights up one wire and one target node. (color and title are required — color must be a CSS variable name like --c-orch, anything else falls back to the brand accent; a missing title shows "—". Color it by the non-orchestrator endpoint: flow-design-patterns.md → "Wire color logic".) A step with from === to renders as a loop on that node — above it, or below for a top-row node (internal work such as "CI → CI: build").
4. Modes — orthogonal toggle (offline/online, dev/prod, v1/v2). Modes can:

• hide/show entire nodes (e.g. Seed only exists offline)
• rename a node (Qdrant → BigQuery)
• swap payload bodies, ports, auth headers, latency chips

Modes are NOT alternative flows. Flows describe scenarios ("user asks a question"); modes describe deployment shape ("on Docker" vs "on Cloud Run").

Workflow when invoked

Step 1 — Capture intent

If the user has uploaded a SOLUTION.md, README, OpenAPI spec, or any architecture description — or described the system in their message — read it first. Extract:

• Service names + tech stack + ports
• Deployment modes (if any)
• Named endpoints/operations and what they do (these become flows)
• For each operation, the chain of internal service calls (these become steps)

Then ask ONLY the gaps that can't be reasonably defaulted. If the system, its main flows and its modes (or lack of them) are known — or the user said not to ask — proceed, and list your assumptions in architecture.md. Defaults: audience = engineers, language = the request's.

Only for a vague request ("diagram our backend") ask briefly: what system, which main flows, any mode toggles (offline/online, dev/prod, v1/v2), who the audience is.

Step 2 — Plan the topology on paper first

Before writing code, sketch on paper (literally in your head or a scratch file) WHICH services exist, WHERE they sit relative to each other, and WHICH flows connect WHICH nodes. Avoid wire crossings — this is the single biggest readability win. See references/flow-design-patterns.md for layout heuristics.

Quick rules:

• Orchestrator/API gateway in the middle, dependencies fanning out
• User on the far left (entry point), datastores on the right (exit), services in between
• Read-only services above, write-heavy services below (intuitive: data flows down into them)
• Optional/one-shot services (seed jobs, cron) tucked in a corner

Step 3 — Build from the template

Copy assets/template.html to the working directory. It's a fully working drop-in: dark theme, side panel, player controls (play/pause/step), keyboard shortcuts, animated SVG packets, mode toggle. Edit only these regions (the template's top EDIT THESE THINGS comment is the authoritative list):

1. `<title>` + `<h1>` — replace {{SYSTEM_NAME}} and the header text. For a Polish (or other non-English) diagram also set <html lang="pl">.
2. `.modepick` buttons — mode toggle labels (or remove the whole .modepick if single-mode; the panel badge and O key adapt). Three or more modes are fine — O cycles through them.
3. `.flowtabs` — one .flowtab per scenario; data-flow must match a flows key. Tab order = keyboard digits 1–9; nothing else to keep in sync.
4. `.node` divs in .stage — one per service: data-id, data-role, style="left/top", label, tech, port. Place them on the grid in references/flow-design-patterns.md (left 2/26/52/78%, top 8/38/70%, ≥ 24% apart in a row) — fixed 150px tiles overlap or clip otherwise.
5. The `flows` object in JS — one entry per scenario, each with an ordered steps[] array. Set state.mode / state.flow to the keys you want shown first (an unknown key falls back to the first button and logs a warning).

Always edit the .legend: one item per role you use, labelled with what that color means in your system (e.g. compute → "Worker"); delete the rest. Recolor :root CSS variables only for a restyle. After editing, search the file for {{[A-Z] (regex) — a correctly filled file has zero matches.

Don't reinvent the CSS, the player, the SVG wire renderer, or the side panel — they all work.

Step 4 — Validate before showing the user

After generating, do this self-check:

• [ ] Open the file mentally: does every flow's step sequence make narrative sense? (Step 1 sets up, last step delivers a result to the user.)
• [ ] Do wires cross unnecessarily? If yes, reposition nodes (try moving the late-arriving dependency away from the early ones).
• [ ] Are there flows where a single node is the to of two consecutive steps? That's fine if it's a loop, but check the description explains why.
• [ ] Mode swap: pick the most different node between modes (usually the database) and verify ALL its mode-dependent fields are filled in (label, tech, port, payload, chips).
• [ ] Did you generate the companion architecture.md? (Required — see Step 5.)

Then run every check in references/validation.md the environment allows: the static check (needs only node); then, with puppeteer-core + Chrome, node <skill-dir>/references/screenshot.js architecture.html /tmp/arch-preview from the user's project (<skill-dir> = this SKILL.md's folder; exit 1 = page error, 2 = lint warnings — non-zero means fix the HTML) and read the canvas PNGs for overlapping or cut-off tiles; without either, window.archValidate() in the browser console ([] = clean).

Step 5 — Output the files

Always two files:

• architecture.html — the interactive diagram
• architecture.md — text description (components, flows step-by-step, mode differences, scenarios for the workshop)

Save to the current working directory. Platform-specific overrides:

• Claude.ai — use /mnt/user-data/outputs/ instead
• Claude Code / Gemini CLI / OpenCode / Copilot CLI — save to the current working directory (default)

Both should be self-sufficient — someone can read the markdown without opening the HTML, and vice versa.

Step 6 — Present the output

Tell the user where the files were saved and how to open them (open architecture.html on macOS, xdg-open architecture.html on Linux, or just open in browser). If present_files is available (Claude.ai), call it to display the HTML first, then the markdown.

File reference

The `examples/*.json` files are planning specs, not drop-in code — JSON outlines you translate into the template's HTML nodes + inline JS flows object. See references/flow-design-patterns.md → "Example JSON → template mapping" for the field-by-field translation.

Quick reference — node roles and colors

Hardcoded in CSS variables in template.html. Pick one per node:

Roles are color groups, not literal meanings; the legend labels say what each color means in your diagram. Don't invent more (Postgres + Redis are both vector, told apart by icon and label) — past seven the legend becomes a colorblind nightmare. An 8th role needs a --c-<role> in the dark :root and body.light blocks, a .node[data-role] rule, an ahMap entry + <marker>, and a legend item.

Visual states — what the viewer sees on each step

This is the most important didactic mechanic in the diagram. Understand it before modifying the JS or CSS, or you'll break the "click a flow tab, see the whole path" intuition.

When a user picks a flow tab and lands on step N, every node and wire is classified into one of these states:

Why `participant` matters (and the JS/CSS implementation gotchas) is detailed in references/flow-design-patterns.md → "The participant state". Short version: it lets a viewer see the whole path on the first click, instead of stepping through to discover which nodes a flow uses — that's the entire point of the diagram. If you add node states, keep applyStep and the CSS (.node.participant, .wire.preview) consistent.

Interaction model

The diagram supports five ways to interact, in increasing order of viewer expertise:

Node click vs drag. A 5px movement threshold distinguishes a click (jump to step) from a drag (reposition). On touch screens nodes are tap-to-jump only, so swiping over a node still scrolls the page.

Node click is the easiest to miss when modifying the skill. Three things must be in place:

1. cursor: pointer on .node (visual affordance — "this is clickable")
2. :hover state that reveals the role color (confirms interactivity)
3. The JS handler that does findIndex on the current flow's steps and jumps there

If a node isn't part of the current flow (.dimmed state), clicking it shakes the node instead of doing nothing. Silent ignore is bad UX — the viewer thinks the page is broken.

Keyboard accessibility: nodes have tabIndex=0, Enter/Space triggers the same action as click, Shift+Arrows moves the node one grid cell. Step changes are announced through a live region. Single-key shortcuts can be switched off via the "shortcuts on" link under the tabs (WCAG 2.1.4).

Don't break this when adding features. It's the only way for a viewer to think "wait, where is the LLM used?" and find out instantly. Without it, they'd have to step through all 8 steps just to discover one node's role.

Quick reference — payload format

Payloads in the side panel are monospace + minimal syntax highlight. Keep them realistic but trimmed — show the shape, omit boring fields:

{
  payloadOffline:
`POST /ask
{
  "query": "Ile kosztuje parking?",
  "limit": 3,
  "temperature": 0.7
}`
  // avoid: full HTTP headers, User-Agent, 50-line bodies — show only the shape
}

Use 4-space indent, no tabs. Use // comments for inline annotations.

If a step is the same in both modes, use payload (single key). If they differ, use payloadOffline + payloadOnline (or whatever the mode names are: payload + capitalized mode key, so mode prod → payloadProd). The mode-specific key wins when both are present — handy for "generic payload, one mode overrides it". Same rule for chips / chipsOffline. An object instead of a string is pretty-printed as JSON.

desc accepts inline HTML only: <em>, <strong>, <code>, <kbd>, <br>, <a href="https://…">. Everything else is stripped at render time (block tags unwrap, scripts/iframes are dropped) — keep descriptions to a sentence or three.

Quick reference — chips (metadata badges)

Chips below the payload are short metadata labels. Two semantic types auto-style themselves:

• Latency (matches /\d+\s*(?:ms|s|min)\b/) — gets a stopwatch icon, e.g. "~150ms", "5-30s", "timeout 120s"
• Size/count (matches /dim|×|loaded|rules|count/) — gets a database icon, e.g. "768 dim", "38 loaded", "×N"
• Other — neutral, e.g. "COSINE", "baseline", "idempotent", "UUID5"

Maximum 3 chips per step. They're hints, not specifications.

Naming conventions

Common pitfalls

Three strings kill the whole diagram. Everything in flows lives inside one <script> block, so in any desc, title, payload or chip:

• </script> closes the block early — write <\/script> (</script> works only in desc; elsewhere it shows literally).
• <!-- with a <script anywhere after it keeps the real </script> from closing the block — write <\!--.
• ${…} in a backtick payload runs as JavaScript (crash, or silently wrong text). Write \${ (e.g. \${{ secrets.TOKEN }}) and ` \ `` for a backtick.

The page lints itself (browser console, window.archValidate(), or screenshot.js): unknown node ids, tab/flow mismatches, unknown onlyMode / data-modes values, missing title/color, leftover placeholders. A clean console is part of "done".

Content Security Policy. The template ships a CSP <meta> that allows only inline script/style and Google Fonts. If a user asks for an external image, script, or font, extend the matching directive (or they'll see a blocked request in the console).

Source node looks dead. If a step's source node renders dimmed instead of active-from, you regressed the participant logic — see "Visual states".

Wire crossings. Caused by node placement, not code — swap node positions to fix. (Wires bend to the right of their direction of travel, so a request and its response sit on opposite sides; repeats between one pair get wider lanes. That helps but won't fix a bad layout.)

Too many flows. More than 6 flows clutters the tab bar. If you have 8+, group them: one diagram per group (e.g. "read flows" vs "write flows" as two separate HTML files).

Modes that aren't really modes. If the toggle changes the entire topology (different services, different protocols), it's not a mode — it's a second system. Build two HTML files.

Static screenshots beat live diagrams sometimes. If the diagram will end up in a PDF report, a slide deck, or a printed handout, this skill is the wrong choice — produce a static diagram instead. This skill is for interactive consumption in a browser.

Forgetting the .md. The HTML is the showpiece, but the .md is what people read on GitHub, paste into Slack, and grep for endpoint names. Always generate both.

Polish vs English. Match the workshop language. Bilingual settings typically use Polish for internal workshops and English for client deliverables. Default to the language of the request; ask only if it's genuinely ambiguous.

Extending the template

If a user asks for something the template doesn't support, here's the difficulty ranking:

Don't promise features the template doesn't support without flagging the work involved.

Iteration tips

When the user gives feedback like "the diagram is too cramped" or "step 5 is confusing":

• "Too cramped" → use the whole grid (Step 3) and shorten tech/port text; never below 24% between nodes in a row. More than ~12 nodes → split the diagram.
• "Confusing step" → the step description is doing too much. Split into two steps OR rewrite to focus on ONE handoff (X sends Y to Z, that's it).
• "It doesn't match how it really works" → ask for the actual code path. Don't guess; the diagram has zero value if it lies about reality.
• "Make it pop more" → resist. Didactic clarity > visual punch. Maybe one accent element (a glow on the result node when flow completes) but stop there.
• "Add a legend" → already there. If they want it bigger, increase the .legend font-size and gap.