XMind Mind Map Processing Skill

.xmind

Tool Script

scripts/xmind_tool.py

python skills/xmind/scripts/xmind_tool.py --session <session-id> <command> [args...]

Session Management

--session <id>

1. Obtain from context: If a previously used session ID already exists in the current conversation context, reuse it directly
2. Generate new ID: If no available session ID exists in the context, generate a UUID v4 format string (e.g.,

   f47ac10b-58cc-4372-a567-0e02b2c3d479

   ) before the first call, and continue to reuse it in subsequent calls

• Memory file storage path:

  /tmp/skills-xmind-parsed/<session-id>/<filename>.md

Supported Formats

content.json

content.xml

Markdown Representation Specification

# Sheet: Worksheet Title

## Central Topic

> Labels: Label1, Label2
> Link: https://example.com
> Markers: priority-1
> This is the note content of the central topic

- Branch 1  {labels: important}  {link: https://example.com}
  > This is the note for Branch 1
  > Notes can have multiple lines
  - Sub-node 1.1  {markers: task-done, priority-1}
    - Sub-sub-node 1.1.1
  - Sub-node 1.2
- Branch 2
  - Sub-node 2.1
  - Sub-node 2.2

• # Sheet: Title

  — Worksheet title (separate multiple worksheets with

  ---

  )

• ## Title

  — Central topic

• - Content

  — Node, use 2-space indentation to indicate hierarchy

• {labels: ...}

  — Inline label metadata

• {link: ...}

  — Inline hyperlink metadata

• {markers: ...}

  — Inline marker/icon metadata

• > Content

  — Node note (in blockquote format, immediately following the node)
• Metadata for the central topic uses top-level blockquotes (

  > Labels:

  ,

  > Link:

  ,

  > Markers:

  )

Core Workflow

Scenario 1: Parse XMind File

1. Confirm File Path: Verify whether the

   .xmind

   file path provided by the user exists
2. Execute Parsing: Run the tool script
   bash

   python skills/xmind/scripts/xmind_tool.py --session <session-id> parse <file.xmind>

3. Display Results: Show the output Markdown content to the user, including:
   • Overview of the overall structure of the mind map
   • Hierarchical relationships of each node
   • Attached metadata such as labels, notes, and links
4. Record Memory Path: The

   <!-- memory_file: ... -->

   comment at the end of the output contains the memory file path; remember this path for subsequent operations
5. Guide Interaction: Prompt the user that they can continue to ask questions, modify, or export the mind map content

Scenario 2: Create XMind File

1. Collect Requirements: Communicate with the user to clarify the structure of the mind map:
   • What is the central topic?
   • What are the main branches?
   • What sub-nodes are under each branch?
   • Do you need to add labels, notes, links, etc.?
2. Generate Markdown: Generate content according to the above Markdown specification based on the user's description
3. Write to Temporary Markdown File and Execute Creation:
   bash

   # Write markdown content to a temporary file, then execute creation
   python skills/xmind/scripts/xmind_tool.py --session <session-id> create <output.xmind> <temp.md> [--format zen|legacy]

4. Confirm Creation Result: Report successful file creation to the user and display the file path
5. Retain Memory: The memory file is automatically created, and can be edited subsequently

• Zen format is created by default (modern format with better compatibility)
• If the user explicitly requires XMind 8 compatible format, use the

  --format legacy

  parameter

Scenario 3: Update XMind File

1. Load Current Content:
   • First attempt to read the memory file (if parsed before):
     bash

     python skills/xmind/scripts/xmind_tool.py --session <session-id> memory <file.xmind>

   • If there is no memory file, perform parsing first:
     bash

     python skills/xmind/scripts/xmind_tool.py --session <session-id> parse <file.xmind>

2. Understand Modification Requirements: Confirm the modifications to be made with the user, such as:
   • Add new branches or sub-nodes
   • Delete a node
   • Modify node content, labels, or notes
   • Adjust node hierarchical relationships
3. Modify Markdown: Make modifications based on the memory file, ensuring the format complies with the specification
4. Write Modified Markdown: Write the modified content to a temporary file
5. Execute Update:
   bash

   python skills/xmind/scripts/xmind_tool.py --session <session-id> update <file.xmind> <modified.md>

6. Confirm Result: Report successful update to the user and display the change summary

Scenario 4: Continue Communication Based on Memory

1. Read Memory File: Use the

   memory

   command to retrieve the previously parsed content
2. Answer Based on Content: Answer the user's question based on the Markdown content
3. Supported Operations:
   • Summarize key points of the mind map
   • Search for specific nodes
   • Analyze structure and hierarchical relationships
   • Propose improvement suggestions
   • Convert content to other formats (outline, table, etc.)

Notes

• Parsing and creation operations involve file I/O; verify the path is valid before execution
• Update operations will overwrite the original file; if the user does not explicitly request it, it is recommended to back up or save as a new file first
• Memory files are stored in

  /tmp/skills-xmind-parsed/<session-id>/

  and may be cleaned up by the system after the session ends
• If a damaged xmind file is encountered (invalid ZIP or missing key content files), report the specific error to the user
• Image attachments in XMind files will not be included in Markdown; only text structure and metadata are processed
• Use Chinese to interact with users throughout the process