Get Glyph
Warning This documentation is still a work in progress. Some details may be out of date depending on the version of Glyph you are using, but it is being actively reviewed and improved.
Documentation AI Assistant Development Licensing

Documentation

Tags & Search

Find notes instantly with full-text search and tag filtering

Glyph provides powerful search capabilities backed by SQLite full-text indexing, making it easy to find any note regardless of how large your space grows.

Access search in multiple ways:

Press Cmd+P (macOS) or Ctrl+P (Windows/Linux) to open the command palette with search.

Click the search icon in the toolbar.

Open the palette and start typing to search.

Search Behavior

When you type a search query:

  1. Debounced: Waits 180ms after you stop typing
  2. Full-text: Searches note titles and content
  3. Ranked results: Most relevant notes appear first
  4. Live updates: Results refresh as you type

Note

Search queries are case-insensitive and support partial word matching.

Search Syntax

Type any word or phrase:

project alpha

Finds notes containing both “project” and “alpha”.

Use quotes for exact phrases:

"user interface design"

Finds notes with that exact phrase.

Keyword Matching

Glyph tokenizes your query and searches for:

  • Individual terms: Each word is a search term
  • Minimum length: Terms must be 2+ characters
  • Stopword filtering: Common words like “the”, “a” may be filtered

Search Results

Results display:

  • Note title: Name of the matching note
  • Snippet: Preview of matching content with highlights
  • Relevance score: Higher scores appear first
  • Match context: Surrounding text for context

View results

Search results appear as you type, ordered by relevance.

Navigate results

Use arrow keys to move between results.

Open note

Press Enter or click a result to open that note.

Ranking Algorithm

Glyph uses a hybrid search algorithm combining:

Keyword Overlap

  • Counts how many search terms appear in the note
  • Higher overlap = higher score

Trigram Similarity

  • Compares character trigrams between query and content
  • Handles typos and fuzzy matching

Phrase Bonus

  • If the exact query phrase appears, boost the score

Title Bonus

  • Notes with query terms in the title rank higher

Formula:

score = (0.6 × overlap) + (0.4 × trigram) + phrase_bonus + title_bonus

Tags

Tag Syntax

Add tags anywhere in your notes:

# Project Notes

#project #important #2024

This note contains three tags.

Tags must:

  • Start with #
  • Contain alphanumeric characters, dashes, underscores
  • Not contain spaces (use - instead: #project-alpha)

Nested Tags

Create hierarchies with /:

#projects/alpha
#projects/beta
#work/client/acme

Tip

Nested tags help organize large tag sets. Searching for #projects can include all nested tags.

Frontmatter Tags

Define tags in YAML frontmatter:

---
tags: [project, draft, urgent]
---

# Note Content

Both inline tags and frontmatter tags are indexed identically.

Filtering by Tag

Use the tag filter in the search interface:

  1. Open search
  2. Select one or more tags
  3. Results narrow to notes with those tags
  4. Optionally add a text query

Tag Autocomplete

When typing tags, Glyph suggests:

  • Existing tags in your space
  • Tag counts (how many notes use each tag)
  • Nested tag paths

Multiple Tags

Select multiple tags to find notes with all selected tags (AND logic):

  • Tag #project AND #urgent
  • Shows only notes containing both tags

Combine text search with tag filtering:

Query: "user interface"
Tags: #project, #draft

Finds notes that:

  1. Contain the phrase “user interface”
  2. Have both #project and #draft tags

Tag Browser

View all tags in your space:

Open Tags view

Click the Tags icon in the sidebar or press the tags shortcut.

Browse tags

See all tags with usage counts.

Click a tag

View all notes with that tag.

The tags list shows:

  • Tag name: The full tag text
  • Count: Number of notes using this tag
  • Sorted: By count (most used first) or alphabetically

Glyph supports advanced search parameters via the Tauri command interface:

Search Options

  • query: Text to search for
  • tags: Array of tags to filter by
  • title_only: Search only note titles
  • tag_only: Search only tags (no content)
  • limit: Maximum results (default: 50)

Example Searches

Search only in note titles, not content:

invoke("search_advanced", {
  request: {
    query: "meeting",
    title_only: true
  }
});

Find notes by tag without text search:

invoke("search_with_tags", {
  tags: ["project", "active"]
});

Text search + tag filter + limit:

invoke("search_advanced", {
  request: {
    query: "design system",
    tags: ["ui", "reference"],
    limit: 20
  }
});

Indexing

Automatic Indexing

Glyph indexes your notes automatically:

  • On save: Every time you save a note
  • On create: When you create a new note
  • On delete: When you delete a note
  • On rename: When you rename a note

Index Contents

The search index stores:

  • Note title: Filename without extension
  • Note content: Full markdown text
  • Tags: Both inline and frontmatter tags
  • Paths: Relative path in your space
  • Timestamps: Created and updated times

Rebuild Index

If search results seem outdated:

Open Settings

Go to Settings > General.

Rebuild Index

Click Rebuild Search Index.

Wait for completion

Glyph re-indexes all notes. Large spaces may take a minute.

Warning

Rebuilding the index locks search temporarily. Avoid doing this during active work sessions.

Recent Notes

View recently modified notes:

invoke("recent_notes", { limit: 10 });

This returns the 10 most recently updated notes, useful for:

  • Quick access to recent work
  • “Continue where you left off” features
  • Recent activity timeline

Performance

Search Speed

Glyph’s search is optimized for spaces with thousands of notes:

  • SQLite FTS5: Full-text search engine
  • Indexed queries: Sub-100ms for most queries
  • Candidate limiting: Fetches top 300 candidates, then ranks
  • Debounced input: Reduces unnecessary queries

Scaling

Search performance characteristics:

Space SizeIndex TimeQuery Time
100 notes< 1s< 10ms
1,000 notes~5s~50ms
10,000 notes~30s~100ms

Tip

Search performance depends on note size. Many small notes search faster than fewer large notes.

Troubleshooting

Search returns no results

Solutions:

  1. Check spelling and try synonyms
  2. Simplify your query (use fewer words)
  3. Rebuild the search index
  4. Verify the note exists in your space

Search is slow

Solutions:

  1. Reduce query complexity
  2. Use more specific search terms
  3. Filter by tags to narrow results
  4. Rebuild index if it’s corrupted

Tags not appearing

Solutions:

  1. Ensure tags start with #
  2. Check for spaces (use #tag-name not #tag name)
  3. Save the note (tags index on save)
  4. Rebuild index if needed