mcpField
O
Obsidian Automation

Obsidian Automation

github.com/claude-office-skillsProductivity5454.8k installsUpdated 39w ago
Repository

About Obsidian Automation

Obsidian Automation is a skill for automating notes and personal knowledge workflows in Obsidian. It covers note templates, automatic linking, Dataview queries, and research workflows.

Key features

  1. 01Create notes from templates
  2. 02Generate links using rules
  3. 03Query tasks and notes with Dataview
  4. 04Automate web clipping and research workflows
  5. 05Analyze note graphs and find connections

Use cases

  1. 01Generate daily notes and standardized notes
  2. 02Automatically link people or projects
  3. 03Review tasks due today and recent meetings
  4. 04Turn web content into research notes

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/claude-office-skills/skills --skill obsidian-automation

After installing, type/obsidian-automationin Claude Code to confirm it loaded.

Details

From the project README

Obsidian Automation

Automate Obsidian knowledge management and personal knowledge base workflows.

Core Capabilities

Note Creation
note_templates:
  daily_note:
    filename: "{{date:YYYY-MM-DD}}"
    folder: "Daily Notes"
    template: |
      # {{date:dddd, MMMM D, YYYY}}
      
      ## Morning Intentions
      - [ ] 
      
      ## Tasks
      - [ ] 
      
      ## Notes
      
      ## Evening Reflection
      
      ---
      [[{{date:YYYY-MM-DD|-1d}}|← Yesterday]] | [[{{date:YYYY-MM-DD|+1d}}|Tomorrow →]]

  meeting_note:
    filename: "Meeting - {{title}} - {{date}}"
    folder: "Meetings"
    template: |
      ---
      date: {{date}}
      attendees: {{attendees}}
      tags: meeting
      ---
      
      # {{title}}
      
      ## Agenda
      
      ## Notes
      
      ## Action Items
      - [ ] 
      
      ## Follow-ups
      
      [[Meetings MOC]]
Smart Linking
auto_linking:
  rules:
    - pattern: "[[Person/{{name}}]]"
      trigger: "@{{name}}"
      create_if_missing: true
      
    - pattern: "[[Project/{{project}}]]"
      trigger: "#proj/{{project}}"
      
  backlink_suggestions:
    enabled: true
    min_mentions: 2
    
  alias_support:
    - "[[Machine Learning|ML]]"
    - "[[Artificial Intelligence|AI]]"
Dataview Queries
dataview_examples:
  tasks_due_today:
    query: |
      ```dataview
      TASK
      WHERE !completed AND due = date(today)
      SORT due ASC
      ```
      
  recent_meetings:
    query: |
      ```dataview
      TABLE date, attendees
      FROM "Meetings"
      WHERE date >= date(today) - dur(7 days)
      SORT date DESC
      LIMIT 10
      ```
      
  project_dashboard:
    query: |
      ```dataview
      TABLE status, due, priority
      FROM #project
      WHERE status != "completed"
      SORT priority ASC
      ```
Templates
templates:
  zettelkasten:
    filename: "{{date:YYYYMMDDHHmmss}}"
    content: |
      ---
      id: {{date:YYYYMMDDHHmmss}}
      tags: 
      links: 
      ---
      
      # {{title}}
      
      ## Idea
      
      ## Source
      
      ## Connections
      - Related to: 
      
      ## References
      
  book_note:
    filename: "Book - {{title}}"
    content: |
      ---
      author: {{author}}
      finished: 
      rating: 
      tags: book
      ---
      
      # {{title}}
      by {{author}}
      
      ## Summary
      
      ## Key Ideas
      
      ## Highlights
      
      ## My Thoughts
      
      ## Action Items

Workflow Automations

Web Clipper
web_clipper:
  trigger: browser_extension
  actions:
    - extract_content:
        title: "{{page.title}}"
        url: "{{page.url}}"
        content: "{{selection}}"
    - create_note:
        folder: "Clippings"
        template: web_clip
    - add_tags: ["web-clip", "{{domain}}"]
Research Workflow
research_workflow:
  steps:
    - create_topic_note:
        filename: "Research - {{topic}}"
        folder: "Research"
    - gather_sources:
        search: "{{topic}}"
        link_to_note: true
    - generate_questions:
        based_on: sources
    - create_sub_notes:
        for_each: key_concept

Graph Analysis

graph_insights:
  orphan_notes:
    query: "notes without incoming links"
    action: suggest_connections
    
  clusters:
    identify: true
    visualize: true
    
  link_suggestions:
    based_on: content_similarity
    threshold: 0.7

Best Practices

  1. Atomic Notes: One idea per note
  2. Consistent Naming: Use conventions
  3. Link Liberally: Connect related ideas
  4. Daily Practice: Regular review
  5. Templates: Standardize note types
  6. Tags vs Links: Use both strategically

Comments

Markdown supported