← Back to Skills Marketplace
0xcjl

Obsidian Viz

by Jialin · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ✓ Security Clean
208
Downloads
2
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install obsidian-viz
Description
Generate Obsidian-compatible visualization files (Excalidraw / Mermaid / Canvas). Supports text descriptions and image inputs, outputs editable diagrams in O...
README (SKILL.md)

Obsidian Viz Skill

Generate Obsidian-compatible visualization files from text descriptions or image inputs.

Processing Flow

Step 0 - Input Type Detection

If user sends an image:

  1. Load modules/image-reader.md
  2. Execute image type recognition and content extraction
  3. Output structured Markdown summary
  4. If image contains diagrams, proceed to Step 1
  5. If image is text/screenshot only, end process

If user provides text description:

  • Proceed directly to Step 1

Step 1 - Tool Selection

Load modules/chart-router.md and select the most appropriate tool based on content type:

  • Excalidraw: Hand-drawn style, architecture diagrams, free layout, concept maps
  • Mermaid: Technical documentation, flowcharts, sequence diagrams, state diagrams, ER diagrams
  • Canvas: Large knowledge networks, interactive exploration, data visualization

Step 2 - Format Specification Loading

Load the corresponding reference file based on the selected tool:

  • mermaidreferences/mermaid.md
  • excalidrawreferences/excalidraw.md
  • canvasreferences/canvas.md

Important: Must read the corresponding reference file before generating any content.

Step 3 - Output Format Selection

Standard Format (when user explicitly requests "standard format" or "excalidraw.com"):

  • Mermaid → .mmd file
  • Excalidraw → .excalidraw file
  • Canvas → .html file

Obsidian Format (default):

  • Mermaid → .md file (with mermaid code block)
  • Excalidraw → .md file (with Excalidraw JSON)
  • Canvas → .canvas file

Step 4 - Generate File

  1. Strictly follow format specifications in reference file
  2. Output to ~/.openclaw/workspace/outputs/\x3Cfilename>.\x3Cext>
  3. Explain to user how to open the file in Obsidian

Usage Instructions

Excalidraw Files

Obsidian Mode (.md):

  • Place in any vault folder
  • Obsidian automatically opens in canvas mode
  • Requires Excalidraw plugin

Standard Mode (.excalidraw):

  • Can be opened directly at excalidraw.com
  • Supports import to any Excalidraw instance

Mermaid Files

Obsidian Mode (.md):

  • Place anywhere in vault
  • Renders in normal preview mode
  • Obsidian supports Mermaid by default

Standard Mode (.mmd):

  • Can be opened in Mermaid-compatible editors
  • Natively supported by GitHub, GitLab, etc.

Canvas Files

Obsidian Mode (.canvas):

  • Place anywhere in vault
  • Double-click to interact in Canvas view
  • Natively supported by Obsidian

Standard Mode (.html):

  • Open in browser
  • Supports interactive exploration

Chart Type Quick Reference

Need Recommended Tool Chart Type
Workflow / CI-CD Excalidraw or Mermaid flowchart
API calls / Message interaction Mermaid sequenceDiagram
Organization / System hierarchy Excalidraw hierarchy
Concept divergence / Brainstorming Canvas or Excalidraw mindmap
State machine / Lifecycle Mermaid stateDiagram-v2
Project timeline Excalidraw timeline
A vs B comparison Excalidraw comparison
Priority matrix Excalidraw matrix
Large knowledge network Canvas free-layout
Animation demo Excalidraw animation mode
Database design Mermaid erDiagram
Class diagram / Object relationships Mermaid classDiagram
Project schedule Mermaid gantt

Notes

  1. Must load reference first: Skipping this step will produce incorrectly formatted files
  2. Chinese support: All tools natively support Chinese, no escaping needed
  3. File path: Output files are uniformly placed in ~/.openclaw/workspace/outputs/ directory
  4. Fallback strategy: If primary tool fails, automatically try alternative tool
  5. Node count limit: For more than 30 nodes, recommend user to split or use Canvas
Usage Guidance
This skill appears coherent and limited to generating Obsidian-compatible diagram files. Before installing or using it: - Be aware it will write output files to ~/.openclaw/workspace/outputs/; inspect those files before copying them into any Obsidian vault. - The skill describes image understanding but contains no OCR/processing binaries — the agent/platform will perform image interpretation; avoid sending sensitive images you wouldn't want processed by the agent. - Generated Canvas files can include 'file' nodes (paths to files in a vault) or 'link' nodes (URLs). Verify any embedded file paths or external links before following them. - Because this is instruction-only, review outputs (especially JSON) for unexpected links or references. If you need higher assurance, test with non-sensitive sample inputs first.
Capability Analysis
Type: OpenClaw Skill Name: obsidian-viz Version: 1.0.2 The obsidian-viz skill bundle is a comprehensive and well-documented tool for generating visualization files (Mermaid, Excalidraw, and Obsidian Canvas) from text or images. The logic is entirely contained within Markdown instructions (SKILL.md and modules/*.md) that guide the AI agent through content extraction and format specification without any evidence of malicious intent, data exfiltration, or unauthorized system access. All file operations are restricted to the standard OpenClaw workspace directory (~/.openclaw/workspace/outputs/).
Capability Assessment
Purpose & Capability
Name, description, and included modules (chart-router, image-reader, and per-tool references) align with the stated goal of generating Excalidraw/Mermaid/Canvas files. There are no unrelated required binaries, env vars, or config paths.
Instruction Scope
SKILL.md instructs the agent to perform image type recognition and content extraction and to strictly follow included format reference files before writing outputs. That scope matches the purpose, but image-processing is only specified at a high level (no implementation code). The skill will read its bundled reference/module markdown files and will write output files to the agent workspace; it may also embed file or link nodes in Canvas/Excalidraw JSON which could reference vault paths or URLs — review generated files before moving them into a personal vault.
Install Mechanism
Instruction-only skill with no install spec and no external downloads. No code is written to disk by an installer; risk from install mechanism is minimal.
Credentials
The skill requests no environment variables, credentials, or config paths. All required inputs are user-provided content (text or images) and the skill writes outputs to the OpenClaw workspace directory.
Persistence & Privilege
The skill is not always-enabled and uses normal autonomous invocation defaults. It does write files to ~/.openclaw/workspace/outputs/ (documented), which is a reasonable scope for an output-generating skill and does not request system-wide configuration changes.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install obsidian-viz
  3. After installation, invoke the skill by name or use /obsidian-viz
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Updated SKILL.md to use concise English phrasing throughout. - Streamlined and clarified instructions, removing duplicated and verbose sections. - Unified terminology for input types, tool selection, and output formats. - Reformatted chart type quick reference for improved readability. - No changes to code or functionality; documentation only.
v1.0.1
Update description to English
v1.0.0
Initial release: Dual-input visualization tool for Obsidian (text description + image understanding)
Metadata
Slug obsidian-viz
Version 1.0.2
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 3
Frequently Asked Questions

What is Obsidian Viz?

Generate Obsidian-compatible visualization files (Excalidraw / Mermaid / Canvas). Supports text descriptions and image inputs, outputs editable diagrams in O... It is an AI Agent Skill for Claude Code / OpenClaw, with 208 downloads so far.

How do I install Obsidian Viz?

Run "/install obsidian-viz" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Obsidian Viz free?

Yes, Obsidian Viz is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Obsidian Viz support?

Obsidian Viz is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Obsidian Viz?

It is built and maintained by Jialin (@0xcjl); the current version is v1.0.2.

💬 Comments