Excalidraw Diagram

Excalidraw Diagram Creator — BidDeed.AI / ZoneWise.AI

Generate .excalidraw JSON files that argue visually, not just display information.

Forked from coleam00/excalidraw-diagram-skill with house brand customizations.

Setup: If the user asks you to set up this skill (renderer, dependencies, etc.), see the Setup section below.

Customization

All colors and brand-specific styles live in one file: references/color-palette.md. Read it before generating any diagram and use it as the single source of truth for all color choices — shape fills, strokes, text colors, evidence artifact backgrounds, everything.

Templates for common patterns live in templates/. Check if a template exists for your diagram type before designing from scratch.

Core Philosophy

Diagrams should ARGUE, not DISPLAY.

A diagram isn't formatted text. It's a visual argument that shows relationships, causality, and flow that words alone can't express. The shape should BE the meaning.

The Isomorphism Test: If you removed all text, would the structure alone communicate the concept? If not, redesign.

The Education Test: Could someone learn something concrete from this diagram, or does it just label boxes? A good diagram teaches—it shows actual formats, real event names, concrete examples.

Brand Defaults

Before generating any diagram, apply these defaults:

Default to light canvas unless the diagram is for investor materials, marketing, or the user requests dark mode.

Templates (Check First)

Before designing from scratch, check if a template exists in templates/:

BidDeed.AI Templates

• templates/biddeed/pipeline-12stage.md — Everest Ascent 12-stage linear flow
• templates/biddeed/bid-decision.md — BID/REVIEW/SKIP decision tree
• templates/biddeed/financial-evidence.md — Bid amounts, ROI, judgment ratios

ZoneWise.AI Templates

• templates/zonewise/waterfall-tier.md — N-tier fallback with cost annotations
• templates/zonewise/multi-repo-map.md — Repo → deploy → datastore network
• templates/zonewise/county-grid.md — FL county status matrix
• templates/zonewise/nightly-pipeline.md — GitHub Actions cron workflow
• templates/zonewise/cost-comparison.md — Before/after side-by-side with $

Shared Templates

• templates/shared/team-org.md — Multi-agent hierarchy
• templates/shared/smart-router.md — Multi-tier LLM routing
• templates/shared/data-flow.md — Input → transform → output with JSON evidence

Templates provide layout guidance and element patterns. Combine with the color palette for final colors.

Depth Assessment (Do This First)

Before designing, determine what level of detail this diagram needs:

Simple/Conceptual Diagrams

Use abstract shapes when:

• Explaining a mental model or philosophy
• The audience doesn't need technical specifics
• The concept IS the abstraction (e.g., "separation of concerns")

Comprehensive/Technical Diagrams

Use concrete examples when:

• Diagramming a real system, protocol, or architecture
• The diagram will be used to teach or explain (e.g., YouTube video)
• The audience needs to understand what things actually look like
• You're showing how multiple technologies integrate

For technical diagrams, you MUST include evidence artifacts (see below).

Research Mandate (For Technical Diagrams)

Before drawing anything technical, research the actual specifications.

If you're diagramming a protocol, API, or framework:

1. Look up the actual JSON/data formats
2. Find the real event names, method names, or API endpoints
3. Understand how the pieces actually connect
4. Use real terminology, not generic placeholders

Bad: "Scraper" → "Database" Good: "Firecrawl ($83/mo) → markdown" → "Gemini Flash (FREE) → structured JSON" → "Supabase multi_county_auctions (245K rows)"

Research makes diagrams accurate AND educational.

Evidence Artifacts

Evidence artifacts are concrete examples that prove your diagram is accurate and help viewers learn. Include them in technical diagrams.

Types of evidence artifacts (choose what's relevant to your diagram):

The key principle: show what things actually look like, not just what they're called.

Multi-Zoom Architecture

Comprehensive diagrams operate at multiple zoom levels simultaneously. Think of it like a map that shows both the country borders AND the street names.

Level 1: Summary Flow

A simplified overview showing the full pipeline or process at a glance. Often placed at the top or bottom of the diagram.

Example: Firecrawl → Gemini → Claude → Supabase or Discovery → Scraping → Analysis → Report

Level 2: Section Boundaries

Labeled regions that group related components. These create visual "rooms" that help viewers understand what belongs together.

Example: Grouping by tier (Tier 1 / Tier 2 / Tier 3), by phase (Scrape / Parse / Store), or by cost (FREE / Paid)

Level 3: Detail Inside Sections

Evidence artifacts, code snippets, cost data, and concrete examples within each section. This is where the educational value lives.

Example: Inside a "Tier 2" section, you might show the actual Gemini Flash API response format, not just a box labeled "AI Parser"

For comprehensive diagrams, aim to include all three levels.

Bad vs Good

Container vs. Free-Floating Text

Not every piece of text needs a shape around it. Default to free-floating text. Add containers only when they serve a purpose.

Typography as hierarchy: Use font size, weight, and color to create visual hierarchy without boxes. A 28px title doesn't need a rectangle around it.

The container test: For each boxed element, ask "Would this work as free-floating text?" If yes, remove the container.

Design Process (Do This BEFORE Generating JSON)

Step 0: Check Templates

Before designing from scratch, check templates/ for an existing pattern that matches your diagram type.

Step 1: Assess Depth Required

Determine if this needs to be:

• Simple/Conceptual: Abstract shapes, labels, relationships
• Comprehensive/Technical: Concrete examples, code snippets, real data, cost annotations

If comprehensive: Do research first. Look up actual specs, formats, event names, APIs.

Step 2: Understand Deeply

Read the content. For each concept, ask:

• What does this concept DO? (not what IS it)
• What relationships exist between concepts?
• What's the core transformation or flow?
• What would someone need to SEE to understand this?
• What does it cost? (if relevant — always include cost data for infrastructure diagrams)

Step 3: Map Concepts to Patterns

For each concept, find the visual pattern that mirrors its behavior:

Step 4: Ensure Variety

For multi-concept diagrams: each major concept must use a different visual pattern. No uniform cards or grids.

Step 5: Sketch the Flow

Before JSON, mentally trace how the eye moves through the diagram. There should be a clear visual story.

Step 6: Generate JSON

Only now create the Excalidraw elements. See below for how to handle large diagrams.

Step 7: Render & Validate (MANDATORY)

After generating the JSON, you MUST run the render-view-fix loop until the diagram looks right. See the Render & Validate section below.

Large / Comprehensive Diagram Strategy

For comprehensive or technical diagrams, you MUST build the JSON one section at a time. Do NOT attempt to generate the entire file in a single pass. This is a hard constraint — Claude Code has a ~32,000 token output limit per response, and a comprehensive diagram easily exceeds that in one shot.

The Section-by-Section Workflow

Phase 1: Build each section

1. Create the base file with the JSON wrapper (type, version, appState, files) and the first section of elements.
2. Add one section per edit. Each section gets its own dedicated pass.
3. Use descriptive string IDs (e.g., "trigger_rect", "arrow_fan_left") so cross-section references are readable.
4. Namespace seeds by section (e.g., section 1 uses 100xxx, section 2 uses 200xxx) to avoid collisions.
5. Update cross-section bindings as you go.

Phase 2: Review the whole

After all sections are in place, check:

• Are cross-section arrows bound correctly on both ends?
• Is the overall spacing balanced?
• Do IDs and bindings all reference elements that actually exist?

Phase 3: Render & validate

Now run the render-view-fix loop from the Render & Validate section.

What NOT to Do

• Don't generate the entire diagram in one response.
• Don't use a coding agent to generate the JSON.
• Don't write a Python generator script.

Visual Pattern Library

Fan-Out (One-to-Many)

Central element with arrows radiating to multiple targets. Use for: sources, orchestrators, root causes, central hubs.

        ○
       ↗
  □ → ○
       ↘
        ○

Convergence (Many-to-One)

Multiple inputs merging through arrows to single output. Use for: aggregation, funnels, synthesis.

  ○ ↘
  ○ → □
  ○ ↗

Waterfall (Tiered Fallback) ★ NEW

Vertical cascade where each tier catches what the previous tier couldn't handle. Use for: scraper pipelines, LLM routing, error handling chains.

  ┌──────────┐
  │  Tier 1  │──→ success
  └────┬─────┘
       ↓ fallback
  ┌──────────┐
  │  Tier 2  │──→ success
  └────┬─────┘
       ↓ fallback
  ┌──────────┐
  │  Tier 3  │──→ success
  └──────────┘

Each tier gets a cost annotation and success rate. Arrows going right = success path. Arrows going down = fallback path.

Tree (Hierarchy)

Parent-child branching with connecting lines and free-floating text. Use for: file systems, org charts, repo maps.

  label
  ├── label
  │   ├── label
  │   └── label
  └── label

Spiral/Cycle (Continuous Loop)

Elements in sequence with arrow returning to start.

  □ → □
  ↑     ↓
  □ ← □

Cloud (Abstract State)

Overlapping ellipses with varied sizes.

Assembly Line (Transformation)

Input → Process Box → Output with clear before/after. Include actual data samples.

  ○○○ → [PROCESS] → □□□
  chaos              order

Side-by-Side (Comparison)

Two parallel structures with visual contrast. Use for: before/after, cost comparison, old vs new.

Gap/Break (Separation)

Visual whitespace or barrier between sections.

Lines as Structure

Use lines (type: line, not arrows) as primary structural elements:

• Timelines: Line + small dots + free-floating labels
• Tree structures: Vertical trunk + horizontal branches + text labels
• Dividers: Thin dashed lines to separate sections
• Flow spines: A central line that elements relate to

Shape Meaning

Rule: Default to no container. Add shapes only when they carry meaning. Aim for <30% of text elements to be inside containers.

Color as Meaning

Colors encode information, not decoration. Every color choice should come from references/color-palette.md.

Key principles:

• Each semantic purpose has a specific fill/stroke pair
• Free-floating text uses color for hierarchy (titles, subtitles, details)
• Evidence artifacts use dark background + colored text
• Cost annotations use orange accent text on dark background
• Status badges use semantic colors (green=active/bid, orange=legacy/review, red=deprecated/skip)
• Always pair a darker stroke with a lighter fill for contrast

Do not invent new colors. If a concept doesn't fit an existing semantic category, use Primary/Neutral or Secondary.

Modern Aesthetics

Roughness

• roughness: 0 — Clean, crisp edges. Default for all BidDeed/ZoneWise diagrams.
• roughness: 1 — Hand-drawn, organic feel. Only if explicitly requested.

Stroke Width

• strokeWidth: 1 — Thin, elegant. Good for lines, dividers.
• strokeWidth: 2 — Standard. Good for shapes and primary arrows.
• strokeWidth: 3 — Bold. Use sparingly for emphasis.

Opacity

Always use opacity: 100 for all elements.

Small Markers Instead of Shapes

Instead of full shapes, use small dots (10-20px ellipses) as:

• Timeline markers
• Bullet points
• Connection nodes
• Visual anchors for free-floating text

Layout Principles

Hierarchy Through Scale

• Hero: 300×150 - visual anchor, most important
• Primary: 180×90
• Secondary: 120×60
• Small: 60×40
• Badge: 60×24

Whitespace = Importance

The most important element has the most empty space around it (200px+).

Flow Direction

Guide the eye: typically left→right or top→bottom for sequences, top→bottom for waterfalls, radial for hub-and-spoke.

Connections Required

Position alone doesn't show relationships. If A relates to B, there must be an arrow.

Text Rules

CRITICAL: The JSON text property contains ONLY readable words.

{
  "id": "myElement1",
  "text": "Start",
  "originalText": "Start"
}

Settings: fontSize: 16, fontFamily: 2, textAlign: "center", verticalAlign: "middle"

JSON Structure

{
  "type": "excalidraw",
  "version": 2,
  "source": "https://excalidraw.com",
  "elements": [...],
  "appState": {
    "viewBackgroundColor": "#ffffff",
    "gridSize": 20
  },
  "files": {}
}

For dark canvas mode, set "viewBackgroundColor": "#020617".

Element Templates

See references/element-templates.md for copy-paste JSON templates. Pull colors from references/color-palette.md based on each element's semantic purpose.

Render & Validate (MANDATORY)

You cannot judge a diagram from JSON alone. After generating or editing the Excalidraw JSON, you MUST render it to PNG, view the image, and fix what you see — in a loop until it's right.

How to Render

# PNG (default)
cd .claude/skills/excalidraw-diagram/references && uv run python render_excalidraw.py <path-to-file.excalidraw>

# SVG export
cd .claude/skills/excalidraw-diagram/references && uv run python render_excalidraw.py <path-to-file.excalidraw> --svg

This outputs a PNG/SVG next to the .excalidraw file. Then use the Read tool on the PNG to actually view it.

The Loop

1. Render & View — Run the render script, then Read the PNG.

2. Audit against your original vision — Compare the rendered result to what you designed:

• Does the visual structure match the conceptual structure?
• Does each section use the intended pattern?
• Does the eye flow through the diagram correctly?
• Is the visual hierarchy correct?
• For technical diagrams: are evidence artifacts readable?
• For dark mode: is all text readable against the dark background?
• Are cost annotations clearly visible and correctly positioned?
• Do status badges render at the right size and position?

3. Check for visual defects:

• Text clipped by or overflowing its container
• Text or shapes overlapping other elements
• Arrows crossing through elements
• Labels floating ambiguously
• Uneven spacing
• Text too small to read
• Overall composition lopsided
• On dark canvas: text blending into background

4. Fix — Edit the JSON. Common fixes:

• Widen containers for clipped text
• Adjust coordinates for spacing
• Add waypoints to arrow points arrays
• Reposition labels closer to their element
• Resize for visual balance
• Switch text colors for dark canvas readability

5. Re-render & re-view

6. Repeat — Typically 2-4 iterations.

When to Stop

The loop is done when:

• The rendered diagram matches the conceptual design
• No text is clipped, overlapping, or unreadable
• Arrows route cleanly
• Spacing is consistent and composition is balanced
• You'd be comfortable showing it to an investor without caveats

First-Time Setup

cd .claude/skills/excalidraw-diagram/references
uv sync
uv run playwright install chromium

Quality Checklist

Depth & Evidence

1. Research done: Did you look up actual specs, formats, costs?
2. Evidence artifacts: Code snippets, JSON examples, cost data present?
3. Multi-zoom: Summary flow + section boundaries + detail?
4. Concrete over abstract: Real content shown, not just labeled boxes?
5. Educational value: Could someone learn something concrete?

Conceptual

6. Isomorphism: Visual structure mirrors concept behavior?
7. Argument: Diagram SHOWS something text alone couldn't?
8. Variety: Each major concept uses a different visual pattern?
9. No uniform containers: No card grids or equal boxes?

Container Discipline

10. Minimal containers: Could any boxed element work as free-floating text?
11. Lines as structure: Tree/timeline patterns using lines + text?
12. Typography hierarchy: Font size and color creating hierarchy?

Structural

13. Connections: Every relationship has an arrow or line
14. Flow: Clear visual path for the eye
15. Hierarchy: Important elements larger/more isolated

Brand Compliance

16. Font: fontFamily: 2 (Helvetica)
17. Roughness: roughness: 0
18. Opacity: opacity: 100
19. Colors: All from color-palette.md
20. Canvas: Correct mode (light or dark) for audience
21. Container ratio: <30% of text inside containers

Visual Validation (Render Required)

22. Rendered to PNG: Visually inspected
23. No text overflow: All text fits
24. No overlapping: Nothing overlaps unintentionally
25. Even spacing: Consistent spacing
26. Arrows correct: Connect to intended elements
27. Readable: Text legible in rendered PNG
28. Balanced: No large voids or overcrowded regions
29. Dark mode check: If dark canvas, all text readable