
excalidraw-studio
Generate Excalidraw diagrams from natural language descriptions. Outputs .excalidraw JSON files openable in Excalidraw. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", "generate an Excalidraw file", "draw an ER diagram", "create a sequence diagram", or "make a class diagram". Supports flowcharts, relationship diagrams, mind maps, architecture, DFD, swimlane, class, sequence, and ER diagrams. Can use icon libraries (AWS, GCP, etc.) when set up. Do NOT use for code architecture analysis (use the architecture skills), Mermaid diagram rendering (use mermaid-studio), or non-visual documentation (use docs-writer).
- Installs
- 244
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/tech-leads-club/agent-skills --skill excalidraw-studio- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads excalidraw-studio automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Excalidraw Studio
Generate Excalidraw-format diagrams from natural language descriptions. Outputs .excalidraw JSON files that can be opened directly in Excalidraw (web, VS Code extension, or Obsidian plugin).
Workflow
UNDERSTAND → CHOOSE TYPE → EXTRACT → GENERATE → SAVE
Step 1: Understand the Request
Analyze the user's description to determine:
- Diagram type — Use the decision matrix below
- Key elements — Entities, steps, concepts, actors
- Relationships — Flow direction, connections, hierarchy
- Complexity — Number of elements (target: under 20 for clarity)
Step 2: Choose the Diagram Type and Visual Mode
Diagram type:
| User Intent | Diagram Type | Keywords |
|---|---|---|
| Process flow, steps | Flowchart | "workflow", "process", "steps" |
| Connections, dependencies | Relationship | "relationship", "connections", "dependencies" |
| Concept hierarchy | Mind Map | "mind map", "concepts", "breakdown" |
| System design | Architecture | "architecture", "system", "components" |
| Data movement | Data Flow (DFD) | "data flow", "data processing" |
| Cross-functional processes | Swimlane | "business process", "swimlane", "actors" |
| Object-oriented design | Class Diagram | "class", "inheritance", "OOP" |
| Interaction sequences | Sequence Diagram | "sequence", "interaction", "messages" |
| Database design | ER Diagram | "database", "entity", "data model" |
Visual mode — decide upfront and apply consistently to all elements:
| Mode | roughness | fontFamily | When to use |
|---|---|---|---|
| Sketch | 1 | 5 | Default — informal, approachable, Excalidraw-native |
| Clean | 0 | 2 | Executive presentations, formal specs |
| Mixed | zones: 0, shapes: 1 | 5 | Architecture diagrams (structural zones + sketchy shapes) |
Step 3: Extract Structured Information
Extract the key components based on diagram type. For each type, identify:
- Nodes/entities — What are the boxes/shapes?
- Connections — What connects to what, and with what label?
- Hierarchy — What contains what, what comes before what?
- Decision points — Where does the flow branch?
For detailed extraction guidelines per diagram type, read references/element-types.md.
Step 4: Generate the Excalidraw JSON
CRITICAL: Read references/excalidraw-schema.md before generating your first diagram. It contains the correct element format, text container model, and binding system.
Key rules for generation:
-
Text inside shapes — Use
boundElementson the shape and a separate text element withcontainerId. Never use alabelshorthand:[ { "id": "step-1", "type": "rectangle", "x": 100, "y": 100, "width": 200, "height": 80, "boundElements": [{ "type": "text", "id": "text-step-1" }] }, { "id": "text-step-1", "type": "text", "x": 130, "y": 128, "width": 140, "height": 24, "text": "My Step", "originalText": "My Step", "fontSize": 20, "fontFamily": 5, "textAlign": "center", "verticalAlign": "middle", "containerId": "step-1", "lineHeight": 1.25, "roundness": null } ] -
Arrow labels — Also use
boundElements+ separate text element withcontainerId. Never use alabelshorthand on arrows:[ { "id": "arrow-1", "type": "arrow", "x": 100, "y": 150, "points": [[0, 0], [200, 0]], "boundElements": [{ "type": "text", "id": "text-arrow-1" }] }, { "id": "text-arrow-1", "type": "text", "x": 160, "y": 132, "width": 80, "height": 18, "text": "sends data", "originalText": "sends data", "fontSize": 14, "fontFamily": 5, "textAlign": "center", "verticalAlign": "middle", "containerId": "arrow-1", "lineHeight": 1.25, "roundness": null } ] -
Arrow bindings — Use
startBinding/endBinding(notstart/end). Connected shapes must list the arrow in theirboundElements:{ "id": "shape-1", "boundElements": [ { "type": "text", "id": "text-shape-1" }, { "type": "arrow", "id": "arrow-1" } ] }{ "id": "arrow-1", "type": "arrow", "startBinding": { "elementId": "shape-1", "focus": 0, "gap": 1 }, "endBinding": { "elementId": "shape-2", "focus": 0, "gap": 1 } } -
Element order for z-index — Always declare shapes first, arrows second, text elements last. This guarantees text renders on top and is never obscured by arrows or other shapes.
-
Positioning — Use grid-aligned coordinates (multiples of 20px when
gridSize: 20). Leave 200-300px horizontal gap, 100-150px vertical gap between elements. -
Unique IDs — Every element must have a unique
id. Use descriptive IDs like"step-1","decision-valid","arrow-1-to-2","text-step-1". -
Colors — Use a consistent palette:
Role Color Hex Primary entities Light blue #a5d8ffProcess steps Light green #b2f2bbImportant/Central Yellow #ffd43bWarnings/Errors Light red #ffc9c9Secondary Cyan #96f2d7Default stroke Dark #1e1e1e
Step 5: Save and Present
-
Save as
<descriptive-name>.excalidraw -
Provide a summary:
Created: user-workflow.excalidraw Type: Flowchart Elements: 7 shapes, 6 arrows, 1 title Total: 14 elements To view: 1. Visit https://excalidraw.com → Open → drag and drop the file 2. Or use the Excalidraw VS Code extension 3. Or open in Obsidian with the Excalidraw plugin
Templates
Pre-built templates are available in assets/ for quick starting points. Use these when the diagram type matches — they provide correct structure and styling:
| Template | File |
|---|---|
| Flowchart | assets/flowchart-template.json |
| Relationship | assets/relationship-template.json |
| Mind Map | assets/mindmap-template.json |
| Data Flow (DFD) | assets/data-flow-diagram-template.json |
| Swimlane | assets/business-flow-swimlane-template.json |
| Class Diagram | assets/class-diagram-template.json |
| Sequence Diagram | assets/sequence-diagram-template.json |
| ER Diagram | assets/er-diagram-template.json |
Read a template when creating that diagram type for the first time. Use its structure as a base, then modify elements to match the user's request.
Icon Libraries
For professional architecture diagrams with service icons (AWS, GCP, Azure, etc.), icon libraries can be set up. Read references/icon-libraries.md when:
- User requests an AWS/cloud architecture diagram
- User mentions wanting specific service icons
- You need to check if icon libraries are available
Best Practices
Element Count
| Diagram Type | Recommended | Maximum |
|---|---|---|
| Flowchart steps | 3-10 | 15 |
| Relationship entities | 3-8 | 12 |
| Mind map branches | 4-6 | 8 |
| Sub-topics per branch | 2-4 | 6 |
If the user's request exceeds maximum, suggest breaking into multiple diagrams:
"Your request includes 15 components. For clarity, I recommend: (1) High-level architecture diagram with 6 main components, (2) Detailed sub-diagrams for each subsystem. Want me to start with the high-level view?"
Layout
- Flow direction: Left-to-right for processes, top-to-bottom for hierarchies
- Spacing: 200-300px horizontal, 100-150px vertical between elements
- Grid alignment: Position on multiples of 20px for clean alignment
- Margins: Minimum 50px from canvas edge
- Text sizing: 28-36px titles, 18-22px labels, 14-16px annotations
- Font: Use
fontFamily: 5(Excalifont) for hand-drawn consistency. Fallback to1(Virgil) if 5 is not supported. - Background zones: For architecture diagrams, add semi-transparent dashed zone rectangles (
opacity: 35,strokeStyle: "dashed",roughness: 0) as the first elements in the array to create visual grouping regions. Seereferences/excalidraw-schema.md→ Background Zones. - Element order: zones first → shapes → arrows → text elements (ensures correct z-index and text always renders on top)
Common Mistakes to Avoid
- ❌ Using
label: { text: "..." }shorthand on shapes or arrows — not supported by the Excalidraw parser - ❌ Putting
textdirectly on shape elements withoutcontainerId - ❌ Using
start/endfor arrow bindings — usestartBinding/endBindingwithelementId/focus/gap - ❌ Forgetting to add arrows to their connected shapes'
boundElementsarrays - ❌ Omitting
originalText,lineHeight,autoResize, orbackgroundColor: "transparent"from text elements inside containers - ❌ Omitting required base properties (
angle,strokeStyle,opacity,groupIds,frameId,index,isDeleted,seed,version,versionNonce,updated,link,locked) — elements will not render - ❌ Missing
"files": {}at the top level of the JSON - ❌ Using
roundness: { "type": 3 }on ellipses — ellipses must useroundness: null - ❌ Missing
lastCommittedPoint,startArrowhead,endArrowheadon arrows - ❌ Declaring text elements before arrows — text renders underneath and gets obscured
- ❌ Floating arrows without bindings (won't move with shapes)
- ❌ Overlapping elements (increase spacing)
- ❌ Inconsistent color usage (define palette upfront)
- ❌ Too many elements on one diagram (break into sub-diagrams)
Validation Checklist
Before delivering the diagram, verify:
- All elements have unique IDs
- Every element has ALL required base properties:
angle,strokeStyle,opacity,groupIds,frameId,index,isDeleted,link,locked,seed,version,versionNonce,updated -
indexvalues are assigned in order ("a0","a1", …) with text elements getting higher values than shapes/arrows - Top-level JSON includes
"files": {} - Shapes with text use
boundElements+ separate text element withcontainerId - Text elements inside containers have
containerId,originalText,lineHeight: 1.25,autoResize: true,roundness: null,backgroundColor: "transparent" - Arrows use
startBinding/endBinding(withelementId,focus,gap) when connecting shapes, pluslastCommittedPoint: null,startArrowhead: null,endArrowhead: "arrow" - Connected shapes list the arrow in their
boundElementsarrays - Element order: shapes → arrows → text elements (text always on top)
- Ellipses use
roundness: null(not{ "type": 3 }) - Coordinates prevent overlapping (check spacing)
- Text is readable (font size 16+)
- Colors follow consistent scheme
- File is valid JSON
- Element count is reasonable (<20 for clarity)
Troubleshooting
| Issue | Solution |
|---|---|
| Text not showing in shapes | Use boundElements + separate text element with containerId, originalText, lineHeight |
| Text hidden behind arrows | Move text elements to end of elements array (after all arrows) |
| Arrows don't move with shapes | Use startBinding/endBinding with elementId, focus: 0, gap: 1 |
| Shape not moving with arrows | Add the arrow to the shape's boundElements array |
| Elements overlap | Increase spacing between coordinates |
| Text doesn't fit | Increase shape width or reduce font size |
| Too many elements | Break into multiple diagrams |
| Colors look inconsistent | Define color palette upfront, apply consistently |
Limitations
- Complex curves are simplified to straight/basic curved lines
- Hand-drawn roughness is set to default (1)
- No embedded images in auto-generation (use icon libraries for service icons)
- Maximum recommended: 20 elements per diagram for clarity
- No automatic collision detection — use spacing guidelines
FAQ
What does the excalidraw-studio skill do?
Generate Excalidraw diagrams from natural language descriptions. Outputs .excalidraw JSON files openable in Excalidraw. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", "generate an Excalidraw file", "draw an ER diagram", "create a sequence diagram", or "make a class diagram". Supports flowcharts, relationship diagrams, mind maps, architecture, DFD, swimlane, class, sequence, and ER diagrams. Can use icon libraries (AWS, GCP, etc.) when set up. Do NOT use for code architecture analysis (use the architecture skills), Mermaid diagram rendering (use mermaid-studio), or non-visual documentation (use docs-writer).
How do I install excalidraw-studio?
Run npx skills add https://github.com/tech-leads-club/agent-skills --skill excalidraw-studio in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/excalidraw-studio/ (personal) or .claude/skills/excalidraw-studio/ (project).
Which AI agents can use excalidraw-studio?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is excalidraw-studio safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains excalidraw-studio?
It is published by tech-leads-club in tech-leads-club/agent-skills. It has 244 recorded installs.
More from tech-leads-club/agent-skills
tlc-spec-driven
tech-leads-club/agent-skills
Feature planning and implementation with 4 adaptive phases (Specify, Design, Tasks, Execute). Auto-sizes depth by complexity. Asks every product decision before the spec is approved, writes testable requirements in EARS notation, atomic tasks with a gate and a commit each, and one check per criterion with the test that proves it. Captures that come with the request bind, and each screen is judged against them. Ships deterministic Python scripts so the gates are enforced by code, not memory; the proofs run once at the end, and an independent Verifier (author != verifier) spends its time on mutations and judgement. Use when (1) planning features, (2) implementing with verification and atomic commits, (3) validating an implementation against a spec. Triggers on "specify feature", "discuss feature", "design", "tasks", "implement", "validate", "verify work", "UAT", "record decision", "pause work", "resume work". Do NOT use for pure architecture decomposition analysis or standalone technical
subagent-creator
tech-leads-club/agent-skills
Guide for creating AI subagents with isolated context for complex multi-step workflows. Use when users want to create a subagent, specialized agent, verifier, debugger, or orchestrator that requires isolated context and deep specialization. Works with any agent that supports subagent delegation. Triggers on "create subagent", "new agent", "specialized assistant", "create verifier". Do NOT use for Cursor-specific subagents (use cursor-subagent-creator instead).
technical-design-doc-creator
tech-leads-club/agent-skills
Creates comprehensive Technical Design Documents (TDD) with mandatory and optional sections through interactive discovery. Use when user asks to "write a design doc", "create a TDD", "technical spec", "architecture document", "RFC", "design proposal", or needs to document a technical decision before implementation. Do NOT use for README files, API docs, or general documentation (use docs-writer instead).
security-best-practices
tech-leads-club/agent-skills
Perform language and framework specific security best-practice reviews and suggest improvements. Use when the user explicitly requests security best practices guidance, a security review or report, or secure-by-default coding help. Supports Python, JavaScript/TypeScript, and Go. Do NOT use for general code review, debugging, threat modeling (use security-threat-model), or non-security tasks.
domain-analysis
tech-leads-club/agent-skills
Maps business domains and suggests service boundaries in any codebase using DDD Strategic Design. Use when asking "what are the domains in this codebase?", "where should I draw service boundaries?", "identify bounded contexts", "classify subdomains", "DDD analysis", or analyzing domain cohesion. Do NOT use for grouping existing components into domains (use domain-identification-grouping) or dependency analysis (use coupling-analysis).
nestjs-modular-monolith
tech-leads-club/agent-skills
Open the skill to read its SKILL.md instructions.
Popular docs & writing skills
grill-with-docs
mattpocock/skills
A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
lark-markdown
open.feishu.cn
飞书 Markdown:查看、创建、上传、编辑和比较飞书中的原生 Markdown 文件。当用户要操作飞书 Markdown 文件,或比较其远端版本及本地草稿时使用。纯本地 Markdown 文件操作不触发本 skill。不负责将 Markdown 导入为飞书在线文档,也不负责文件搜索、权限、评论、移动、删除等云空间管理操作。
writing-beats
mattpocock/skills
Writing, exploit; assemble raw material into a journey of beats, grounding each term before a beat leans on it.
writing-shape
mattpocock/skills
Writing, exploit: shape raw material into an article, paragraph by paragraph.
writing-fragments
mattpocock/skills
Writing, explore: mine raw fragments, no structure yet.
writing-for-agents
mattpocock/skills
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.