Tags

Tags provide a flexible way to organize and categorize objects across your vault. Extenote supports both simple tags and hierarchical tags with a colon separator.

Tag Format

Tags are defined in the frontmatter tags field:

---
title: My Research Paper
tags:
  - research
  - ml
  - collection:data-leverage
---

Simple Tags

Simple tags are single labels like research, important, or draft.

Hierarchical Tags

Hierarchical tags use a colon separator to create parent-child relationships:

• collection:data-leverage - belongs to the “data-leverage” subcollection
• status:draft - has “draft” status
• topic:ml:transformers - deeply nested topic

Hierarchical tags are displayed as a tree structure in the web UI, with the prefix as the parent node.

Web UI Tag Management

The web app provides a full-featured tag management interface at the Tags page.

Features

1. Tag Tree View

   • Hierarchical display with expand/collapse
   • Object counts for each tag
   • Search/filter to find specific tags

2. Tag Operations

   • Rename: Change a tag name across all objects
   • Merge: Combine one tag into another
   • Delete: Remove a tag from all objects

3. Preview Before Apply

   • See exactly which files will be modified
   • Review current vs new tags for each file
   • Cancel if the changes aren’t what you expected

CLI Tag Management

Tag operations are also available via the CLI:

# List all tags with counts
bun run cli -- tags list

# List tags in a specific project
bun run cli -- tags list shared-references

# List as JSON
bun run cli -- tags list --json

# Rename a tag
bun run cli -- tags rename old-tag new-tag

# Preview rename without applying
bun run cli -- tags rename old-tag new-tag --dry-run

# Merge source tag into target
bun run cli -- tags merge source-tag target-tag

# Delete a tag from all objects
bun run cli -- tags delete unwanted-tag

# Delete with confirmation prompt skipped
bun run cli -- tags delete unwanted-tag --yes

Collection Tags

Tags prefixed with collection: define which paper collections an object belongs to:

tags:
  - collection:data-leverage
  - collection:data-valuation

A paper can belong to multiple collections by having multiple collection:* tags.

Priority in Collections

To manually control the order of papers in collections, add a priority:<number> tag:

tags:
  - collection:data-valuation
  - priority:1  # Appears first in all collections this paper belongs to

Sort order: Papers with priority:N tags are sorted by N (ascending), then unprioritized papers are sorted by year (most recent first).

Semble Collection Integration

Tags prefixed with collection: have special meaning for Semble sync. When syncing objects to Semble:

1. The project name becomes a Semble collection
2. Each collection:* tag creates an additional collection membership

For example, an object with:

tags:
  - collection:data-leverage
  - collection:papers

In project shared-references will be synced to three Semble collections:

• shared-references (project collection)
• shared-references:data-leverage (tag collection)
• shared-references:papers (tag collection)

Best Practices

1. Use consistent naming conventions

   • Lowercase with hyphens: machine-learning, data-leverage
   • Avoid spaces or special characters

2. Use hierarchical tags for categories

   • status:draft, status:published, status:archived
   • topic:ml, topic:nlp, topic:vision

3. Preview before applying

   • Always use --dry-run in CLI or Preview in web UI
   • Verify the affected files are what you expect

4. Merge rather than delete

   • When consolidating tags, merge into the canonical name
   • This preserves the tagging rather than losing it

5. Use collection: prefix for Semble

   • Objects with collection:* tags get multi-collection membership
   • Plan your collection structure before syncing