Stats
Actions
Tags
From cavepaintings
Use when the user wants to draw diagrams, annotate screenshots, sketch architecture, or send visual content to this conversation. Also use when the user asks you to draw, sketch, or place objects on a canvas. Triggers on mentions of drawing, sketching, diagramming, whiteboard, visual canvas, or annotating images.
How this skill is triggered — by the user, by Claude, or both
Slash command
/cavepaintings:cavepaintingsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
1. Start (or reuse) the server:
Start (or reuse) the server:
CAVEPAINTINGS_PROJECT_DIR="$(basename "$(pwd)")" <SKILL_DIR>/scripts/start-server.sh
This checks for an existing session automatically. If one is running, it returns the existing connection info with "status": "existing". Otherwise it starts a new server. Either way, save the url and port from the JSON response.
Tell the user:
"Cavepaintings canvas open at {url}. Draw diagrams, paste images, and click 'Submit to Claude' (or Ctrl+Enter) to send them here."
Receiving submissions: The UserPromptSubmit hook automatically detects new submissions when the user sends any message. It surfaces all unseen submissions with prompt text, object summaries, and PNG/JSON file paths. You can also check manually:
node <SKILL_DIR>/scripts/check-session.js --submissions
Read the PNG with the Read tool to see the visual content.
Pushing objects to the canvas: POST Fabric.js objects to http://localhost:{port}/api/canvas with {"diagram":{"objects":[...]},"mode":"merge"}. Objects appear on the canvas in real time. Use "merge" to add or "replace" to clear and load.
Polling submissions via API: GET http://localhost:{port}/api/submissions?since=0 returns all submissions newer than the given timestamp.
POST /api/canvas) only works while a browser tab is connectednpx claudepluginhub sds-mode/cavepaintings --plugin cavepaintingsManages Obsidian canvas files as visual reference boards: creates canvases, adds images/text/PDFs/wiki pages with auto-positioning into zones, lists contents, and integrates with image capture.
AI-orchestrated visual production for Obsidian Canvas. Create presentations, flowcharts, mood boards, knowledge graphs, galleries, storyboards, timelines, dashboards, and more with intelligent layout and AI-generated content. Claude acts as Creative Director — dispatching sub-agents for image generation, SVG diagrams, GIF creation, and spatial layout. Supports 12 template archetypes, 6 layout algorithms, and Advanced Canvas presentation mode. Triggers on: /canvas, create canvas, build canvas, make a presentation, visual board, mood board, flowchart canvas, storyboard, canvas from template, lay out canvas, export canvas, canvas layout, canvas generate, add to canvas, put this on the canvas, open canvas, canvas present, canvas template.
Generates architecture diagrams on a live Excalidraw canvas from text, components, or samples for data flows, call chains, and exports to PNG/SVG/Excalidraw.