Excalidraw logo

Excalidraw

Organization
neo4j-labs
excalidraw

Generate Excalidraw JSON format diagrams for documentation, architecture diagrams, flowcharts, entity relationships, and visual explanations. Use when users request Excalidraw diagrams, .excalidraw files, visual diagrams for docs, architecture visualizations, flowcharts, or when converting ASCII art/placeholder diagrams to proper Excalidraw format.

Overview

Publisherneo4j-labs
Repositoryagent-memory
Skill nameexcalidraw
Stars
585
Forks
108
Bundled files
1
LicenseApache-2.0
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.

  • 1 bundled files

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

  • Open source

    Published by neo4j-labs on GitHub. Read the source before you install it.

Installation

Install the Excalidraw 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/neo4j-labs/agent-memory/tree/main/.claude/skills/docs-excalidraw
  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/neo4j-labs/agent-memory.git /tmp/agent-memory
mkdir -p .claude/skills
cp -r /tmp/agent-memory/.claude/skills/docs-excalidraw .claude/skills/excalidraw
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Excalidraw 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 Excalidraw 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 Excalidraw 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.

Excalidraw Diagram Generator

Generate valid Excalidraw JSON for diagrams that can be opened in Excalidraw, embedded in documentation, or used in Obsidian.

JSON Schema Structure

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

Element Types

Base Properties (all elements)

json
{
  "id": "unique-id",
  "type": "rectangle|ellipse|diamond|text|arrow|line",
  "x": 0,
  "y": 0,
  "width": 100,
  "height": 50,
  "angle": 0,
  "strokeColor": "#1e1e1e",
  "backgroundColor": "transparent",
  "fillStyle": "solid",
  "strokeWidth": 2,
  "strokeStyle": "solid",
  "roughness": 1,
  "opacity": 100,
  "seed": 12345,
  "version": 1,
  "isDeleted": false,
  "groupIds": [],
  "frameId": null,
  "roundness": { "type": 3 },
  "boundElements": null
}

Shape Elements: rectangle, ellipse, diamond

Use for boxes, containers, nodes. Diamond for decision points.

Text Elements

json
{
  "type": "text",
  "text": "Label",
  "fontSize": 20,
  "fontFamily": 1,
  "textAlign": "center",
  "verticalAlign": "middle",
  "containerId": null
}

fontFamily: 1=Virgil (hand-drawn), 2=Helvetica, 3=Cascadia (code)

Linear Elements: arrow, line

json
{
  "type": "arrow",
  "points": [[0, 0], [100, 0]],
  "startBinding": null,
  "endBinding": null,
  "startArrowhead": null,
  "endArrowhead": "arrow"
}

startArrowhead/endArrowhead: null, "arrow", "bar", "dot", "triangle"

Binding Arrows to Shapes

To connect arrows to shapes, use bindings:

json
{
  "startBinding": {
    "elementId": "target-shape-id",
    "focus": 0,
    "gap": 5
  },
  "endBinding": {
    "elementId": "target-shape-id",
    "focus": 0,
    "gap": 5
  }
}

When binding, add boundElements to the target shape:

json
{
  "boundElements": [
    { "id": "arrow-id", "type": "arrow" }
  ]
}

Style Values

PropertyValues
fillStyle"solid", "hachure", "cross-hatch"
strokeStyle"solid", "dashed", "dotted"
roughness0 (architect), 1 (artist), 2 (cartoonist)
roundness.type2 (small radius), 3 (adaptive radius)

Color Palette

Stroke: #1e1e1e (black), #e03131 (red), #2f9e44 (green), #1971c2 (blue)
Background: transparent, #ffc9c9 (light red), #b2f2bb (light green), 
            #a5d8ff (light blue), #ffec99 (light yellow), #d0bfff (light purple)

Generation Workflow

  1. Plan layout: Sketch positions on a grid (increments of 50-100px)
  2. Create shapes first: Generate all boxes/nodes with unique IDs
  3. Add text: Create text elements, optionally bound to containers
  4. Add connections: Create arrows with bindings to shape IDs
  5. Update boundElements: Add arrow references to connected shapes

ID Generation

Use descriptive IDs: "memory-client-box", "arrow-client-to-store", "label-extraction"

Common Patterns

Architecture Box with Label

json
[
  {
    "id": "box-1",
    "type": "rectangle",
    "x": 100, "y": 100,
    "width": 200, "height": 80,
    "strokeColor": "#1e1e1e",
    "backgroundColor": "#a5d8ff",
    "fillStyle": "solid",
    "boundElements": [{ "id": "text-1", "type": "text" }]
  },
  {
    "id": "text-1",
    "type": "text",
    "x": 200, "y": 140,
    "text": "Component",
    "fontSize": 20,
    "textAlign": "center",
    "containerId": "box-1"
  }
]

Connected Flowchart

See references/examples.md for complete flowchart and architecture examples.

Output

Save as .excalidraw file (JSON with .excalidraw extension). Present to user for download.

Replacing ASCII Placeholder Diagrams

When converting asciidoc placeholder diagrams like:

+------------------+
|   Component      |
+------------------+

Map to proper Excalidraw elements with appropriate sizing, colors, and connections.

Documentation Diagram Workflow

This project uses a structured workflow for managing documentation diagrams:

Finding Placeholders

Placeholder diagrams in AsciiDoc files use this pattern:

asciidoc
.Diagram Title
[cols="1", options="header"]
|===
| [DIAGRAM PLACEHOLDER: Diagram Name]

a|
[source,text]
----
ASCII art representation here
----
|===

Using the Management Script

bash
# List all placeholder diagrams
make docs-diagrams-list

# Check active image provenance, source bindings and export freshness
make docs-diagrams-status

# Show missing or stale published diagrams and draft sources
make docs-diagrams-missing

# Add image references after generating diagrams
# (review the inserted caption and alt text, and add the manifest entry)
make docs-diagrams-add-refs

File Locations

  • Excalidraw JSON: docs/assets/diagrams/excalidraw/{slug}.excalidraw
  • Exported SVG: docs/modules/ROOT/images/diagrams/{slug}.svg (diagram exports are SVG; only literal UI screenshots stay PNG)
  • Diagram manifest: docs/diagrams/manifest.json
  • Management script: scripts/manage_diagrams.py

Generating or updating a diagram

  1. Read docs/MAINTAINING.md and run make docs-diagrams-status to inspect active image mappings. A matching filename alone is not provenance.

  2. Verify every schema name, operation, threshold and caption against current source. Distinguish application examples, the Python Bolt backend and NAMS.

  3. Save editable JSON to docs/assets/diagrams/excalidraw/{slug}.excalidraw; preserve IDs and update both ends of bindings.

  4. Install optional tools with npm ci --prefix docs/diagrams and npx --prefix docs/diagrams playwright install chromium.

  5. Export a new or changed diagram with node scripts/export_diagrams.mjs <source.excalidraw> docs/modules/ROOT/images/diagrams/{slug}.svg, then record it in docs/diagrams/manifest.json (step 7). node scripts/export_diagrams.mjs --all only re-exports the SVGs the manifest already records, so it never exports a brand-new diagram. Add --png to also write a sibling PNG; that PNG then needs its own manifest record. The exporter renders through Excalidraw itself; pages embed the SVG.

  6. Inspect the full image and its rendered page at desktop and narrow widths. Check legibility, clipping, arrow endpoints, labels and full alt text. The exporter does not certify semantic correctness.

  7. Record source/output SHA-256 values, referring pages, owner and visual review date in docs/diagrams/manifest.json only after review. Every export needs a manifest entry in the same change, even before a page embeds it ("status": "pending-placement"); make docs-diagrams-status fails on an untracked published image. A captured screenshot can have no editable source if its provenance exception is explained.

  8. Embed the SVG with a caption line directly above the image macro, then update the relevant page and run make docs-lint:

    asciidoc
    .Caption sentence that titles the diagram
    image::diagrams/{slug}.svg["Complete alt text that describes the diagram, with commas quoted",width=100%,link=self]

    The caption is the visible figure title; the alt text describes what the diagram shows and does not repeat the caption.

Do not edit generated OpenWiki diagrams or generated API pages. Preserve historical editable scenes in the canonical directory; the active manifest identifies current published uses. See docs/diagrams/source-migrations.json for the migration from former directories.

Diagram Style Guidelines for This Project

  • Memory types: Use distinct colors per memory type

    • Short-Term: Green (#b2f2bb / #2f9e44)
    • Long-Term: Yellow (#ffec99 / #f08c00)
    • Reasoning: Purple (#d0bfff / #9c36b5)
    • Neo4j/Storage: Blue (#a5d8ff / #1971c2)
  • Pipeline stages: Use vertical flow with colored boxes

  • Graph schemas: Use ellipses for nodes, labeled arrows for relationships

  • Architecture: Use rectangles with hierarchy (top-level → components → storage)

The official Neo4j Labs UI takes precedence over a generic palette or font suggestion. Use the project policy in docs/MAINTAINING.md; do not replace it with a custom product logo.

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 Excalidraw AI skill do?

Generate Excalidraw JSON format diagrams for documentation, architecture diagrams, flowcharts, entity relationships, and visual explanations. Use when users request Excalidraw diagrams, .excalidraw files, visual diagrams for docs, architecture visualizations, flowcharts, or when converting ASCII art/placeholder diagrams to proper Excalidraw format.

Why use Excalidraw on TypingMind?

Because you install it once and use it with any model. Excalidraw 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 Excalidraw in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/neo4j-labs/agent-memory/tree/main/.claude/skills/docs-excalidraw. 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 Excalidraw?

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 Excalidraw?

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

Is the Excalidraw AI skill free?

Yes. It is published on GitHub by neo4j-labs under the Apache-2.0 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 👇