Analyze and Query XMind Mind Maps
XMind Skill lets Claude create and read XMind mind maps, Gantt charts, and flowcharts with zero npm dependencies.
Why it matters
Leverage the Model Context Protocol to analyze and query XMind mind maps. Extract, search, and analyze mind map content for research and task management.
Outcomes
What it gets done
Parse and analyze XMind files, extracting complete mind map structures.
Extract and analyze TODO tasks, including context and hierarchy.
Perform intelligent fuzzy searches and extract specific nodes by path or ID.
Scan directories for XMind files and perform multi-file analysis.
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/vb-xmind | bash Capabilities
Tools your agent gets
Parse and analyze XMind files, extract complete mind map structure
Extract and analyze TODO tasks, including task context and hierarchy
Recursively scan XMind files, filter and organize results
Process multiple files simultaneously, compare and analyze across files
Search files by name patterns, recursive directory scanning
Intelligent fuzzy search by paths, ranked search results, full subtree extraction
Direct node access by ID, fast and precise retrieval
Multi-criteria content search, customizable search fields
Overview
XMind MCP Server
XMind Skill lets Claude create and read XMind mind map files - nested topics, notes, Gantt-style project tasks, and flowcharts - supporting both modern and legacy XMind formats with bilingual natural-language input. Use it when you want Claude to generate or analyze actual .xmind files for mind maps, project plans, or flowcharts, in either Chinese or English requests.
What it does
XMind Skill is a Claude Code / Claude Desktop skill for creating and reading XMind mind map files, built with zero npm dependencies using only Node.js built-ins. It can create nested topics, notes (plain and HTML), labels, markers, callouts, boundaries, summaries, relationships, internal links, simple todo/done tasks, and planned tasks with Gantt-style support (dates, duration, progress, priority, dependencies), plus free-positioning flow and logic diagrams. It reads XMind files with full structure parsing, fuzzy search, task tracking, and multi-file/directory analysis, supporting both modern (XMind Zen/2020+) and legacy (XMind 8) formats.
When to use - and when NOT to
Use this when you want Claude to generate a mind map, project plan with Gantt-style tasks, or flowchart as an actual .xmind file, or to read and analyze existing XMind files - searching across them, extracting nodes, or tracking task status. It works with natural-language requests in both Chinese and English. It is not useful if you don't use XMind as your mind-mapping tool, and complex diagram layouts still need to be expressed through the skill's documented JSON input format described in SKILL.md.
Capabilities
- Create XMind files: nested topics, notes, labels, markers, callouts, boundaries, summaries, relationships, internal links, layout structures, simple and Gantt-style planned tasks, file attachments, and free-positioning diagrams.
- Read XMind files: full structure parsing, fuzzy search, task tracking, multi-file analysis, directory scanning, and node extraction.
- Supports both modern (XMind Zen/2020+) and legacy (XMind 8) file formats for reading and writing.
- Bilingual natural-language routing in Chinese and English.
How to install
For Claude Code, symlink or copy the skill into your project or global skills directory:
ln -s /path/to/mcp-xmind/skills/xmind .claude/skills/xmind
For Claude Desktop, build the skill package and upload it:
npm run build:skill
# -> build/xmind-skill.zip
Then go to Claude Desktop > Settings > Capabilities > Skills and upload build/xmind-skill.zip.
Who it's for
Anyone using XMind for mind maps, project plans, or flowcharts who wants Claude to generate or analyze those files directly through natural-language requests instead of building them manually in the XMind app.
Source README
XMind Skill
A Claude Code / Claude Desktop skill for creating and reading XMind mind maps. Zero npm dependencies - uses only Node.js built-ins.
Features
- Create XMind files: nested topics, notes (plain + HTML), labels, markers, callouts, boundaries, summaries, relationships, internal links, layout structures, simple tasks (todo/done), planned tasks with Gantt support (dates, duration, progress, priority, dependencies), file attachments, free-positioning flow/logic diagrams
- Read XMind files: parse full structure, fuzzy search, task tracking, multi-file analysis, directory scanning, node extraction
- Format support: modern (XMind Zen/2020+) and legacy (XMind 8) for both read and write
- Bilingual: natural Chinese and English interaction routing
Install
Claude Code (CLI)
Symlink or copy the skill into your project or global skills directory:
# Per project
ln -s /path/to/mcp-xmind/skills/xmind .claude/skills/xmind
# All projects
ln -s /path/to/mcp-xmind/skills/xmind ~/.claude/skills/xmind
Claude Desktop
Build the skill ZIP and upload it:
npm run build:skill
# → build/xmind-skill.zip
Open Claude Desktop > Settings > Capabilities > Skills > Upload build/xmind-skill.zip.
Usage
Once installed, just ask Claude in natural language:
- "帮我做个思维导图" / "create a mind map"
- "打开这个xmind文件" / "read this xmind"
- "做个项目计划" / "make a Gantt chart"
- "画个流程图" / "draw a flowchart"
See skills/xmind/SKILL.md for the full JSON input format and all supported features.
Quick test
# Create
echo '{"path":"/tmp/test.xmind","sheets":[{"title":"Test","rootTopic":{"title":"Hello"}}]}' | node skills/xmind/scripts/create_xmind.mjs
# Read
echo '{"action":"read","path":"/tmp/test.xmind"}' | node skills/xmind/scripts/read_xmind.mjs
Project structure
skills/xmind/
SKILL.md # Skill definition (routing, format docs, rules)
scripts/
create_xmind.mjs # Create .xmind files (zero deps)
read_xmind.mjs # Read/search/analyze .xmind files (zero deps)
scripts/
build-skill.sh # Package skill into build/xmind-skill.zip
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.