Skip to content

Usage Examples

This guide shows practical examples of using obx with AI assistants.

You: “Set up my day - create today’s daily note, show my open tasks, and list what I worked on yesterday”

The AI will:

  1. Create/open today’s daily note
  2. Use manage-tasks with action: "list" and status: "open"
  3. Read yesterday’s daily note for context

You: “Help me wrap up - show tasks I completed today, move any unfinished ones to tomorrow, and summarize what I accomplished”

The AI will:

  1. List today’s completed tasks
  2. Identify incomplete tasks
  3. Append them to tomorrow’s daily note
  4. Generate a summary for today’s note

You: “Start my weekly review - show all tasks completed this week, any open items, and notes I created”

The AI will:

  1. Open/create the weekly note
  2. List daily notes from the past 7 days
  3. Aggregate completed tasks
  4. Find notes created this week
  5. Summarize in the weekly note

You: “Add to my inbox: Research GraphQL vs REST for the new API”

→ manage-notes action: "append", path: "inbox.md"
→ content: "- [ ] Research GraphQL vs REST for the new API"

You: “Create a new project note for ‘Mobile App Redesign’ with high priority, status planning, and due date March 15”

The AI will:

  1. Create the note with manage-notes action: "write"
  2. Set frontmatter with manage-frontmatter action: "set":
    • priority: high
    • status: planning
    • due: 2024-03-15

You: “Find my note about the Q4 budget and update the status to approved”

1. search-vault action: "search", query: "Q4 budget"
2. manage-frontmatter action: "set", path: "found/note.md", key: "status", value: "approved"

You: “What do I know about machine learning? Show related notes and any gaps”

The AI will:

  1. search-vault action: "search", query: “machine learning”
  2. search-vault action: "tags", tags: “machine-learning”
  3. manage-links action: "backlinks" on main ML note
  4. analyze-vault action: "orphan-notes" to check for isolated ML content
  5. manage-links action: "suggest" to find potential connections

You: “Create a MOC for my programming notes and connect it to my main index”

1. manage-mocs action: "generate", directory: "programming", output: "programming/Programming MOC.md"
2. manage-notes action: "append" — add link to main index
3. bulk-operations action: "tag" — add #moc tag to the MOC

You: “What concepts in my vault need more development?”

1. analyze-vault action: "find-stubs", max_words: 50
2. analyze-vault action: "broken-links" — find mentioned-but-not-created notes
3. analyze-vault action: "orphan-notes" — find disconnected content

You: “Show me all open tasks grouped by priority”

The AI will use manage-tasks action: "list" and organize results by priority tags or frontmatter.


You: “Find all tasks older than 30 days and mark them for review”

1. manage-tasks action: "list", status: "open"
2. Filter by date in task text or note modified date
3. bulk-operations action: "tag" — add #needs-review to affected notes

You: “Show all tasks in my ‘Website Redesign’ project”

1. search-vault action: "search", directory: "projects/Website Redesign"
2. manage-tasks action: "list", directory: "projects/Website Redesign"

You: “The ‘misc’ folder is getting cluttered. Help me organize it by topic”

The AI might:

  1. List notes in misc/
  2. Analyze content and tags
  3. Suggest categorization
  4. Create new folders
  5. bulk-operations action: "move" notes to appropriate locations

You: “I have tags #todo, #TODO, and #to-do. Standardize them all to #todo”

1. search-vault action: "tags" — search each variant
2. For each note: Update tag format
3. Report: "Standardized 23 notes to use #todo"

You: “Archive all notes in projects/ that haven’t been modified in 6 months”

1. analyze-vault action: "find-outdated", days: 180, directory: "projects"
2. bulk-operations action: "move", destination: "archive/projects"
3. bulk-operations action: "set-frontmatter", key: "status", value: "archived"
4. bulk-operations action: "tag", tag: "#archived"

You: “Prepare for my meeting with the design team - create a note from the meeting template and pull in context from recent design discussions”

1. manage-templates action: "apply", template: "meeting", path: "meetings/Design Team 2024-01-15.md"
2. search-vault action: "search", query: "design"
3. manage-notes action: "append" — add relevant context to meeting note

You: “Extract action items from today’s meeting notes and create tasks”

The AI will:

  1. Read the meeting note
  2. Identify action items (lines starting with ”- [ ]” or “Action:”)
  3. Optionally distribute to relevant project notes

You: “Show me all draft articles ready for review”

search-vault action: "frontmatter", query: "status=draft"
→ Filter by: word count > 500, has all required sections
→ Present list for review

You: “Are there any notes that mention ‘API’ but don’t link to my API documentation?”

1. search-vault: "API"
2. For each result: check if it links to API-docs.md
3. Report unlinked mentions
4. Optionally: Add links

You: “Run a full health check on my vault”

The AI will:

  1. analyze-vault action: "stats" — overall metrics
  2. analyze-vault action: "orphan-notes" — disconnected content
  3. analyze-vault action: "broken-links" — missing link targets
  4. analyze-vault action: "find-stubs" — underdeveloped notes
  5. analyze-vault action: "find-outdated" — stale content
  6. Report findings and suggest improvements

Start broad, then narrow:

You: "What notes do I have about Python?"
AI: Shows 45 notes
You: "Just the ones about async programming"
AI: Filters to 8 notes
You: "Which of these are tutorials?"
AI: Identifies 3 tutorials
You: "Add #python-async tag to all of them"
AI: Tags 3 notes

Describe the full workflow:

“Find all notes tagged #book, check which ones don’t have a rating in frontmatter, and add rating=unrated to those”

The AI handles the multi-step process automatically.

Ask open-ended questions:

“What are the most connected notes in my vault?” “Which topics do I write about most?” “What notes should probably be linked together?”