Squire

Obsidian Squire 🛡️

Your faithful note-taking companion. Squire automates index numbering and surfaces related notes so you can stay in your flow state. No manual calculation, no hunting for connections.

✨ Features

• Smart Index Numbering: Automatically generate the new note's prefix (incrementing or nesting numbers) while leaving the title blank for you to instantly type the new subject. Right-click or command palette to duplicate.
• Related Notes: Automatically surface connected notes via word, tag, and link similarity, with an optional semantic embedding engine for deeper content understanding — no manual searching required.

---

🎯 Why Squire? (The Problem)

Every action that breaks your typing flow — calculating the next index number, manually hunting for related notes — adds micro-friction that pulls you out of deep work.

If you use a structural note-taking framework like the Zettelkasten Method, you likely rely on a strict naming or numbering convention. As your vault grows, manually copying, pasting, and calculating the next index number becomes tedious.

Squire removes this friction on two fronts: it automates index numbering so you never do mental arithmetic, and it surfaces related notes so you never hunt for connections.

Example Workflow

Imagine you use a simple numeric index for your notes:

📂 Literature_Notes
 └── 📄 1 - Biology.md

If you want to create a new, unrelated note (e.g., Physics), you need to increment the number:

📂 Literature_Notes
 ├── 📄 1 - Biology.md
 └── 📄 2 - Physics.md

If you want to create a child note related to Biology (e.g., Anatomy), you need to nest the number:

📂 Literature_Notes
 ├── 📄 1 - Biology.md
 └── 📄 1.1 - Anatomy.md

🚀 Smart Index Numbering

No mental arithmetic. No manual formatting. Just pick the command and type your subject.

Right-click a note in the File Explorer and select Duplicate: Increment Last Number or Duplicate: Nest Last Number, or open the Command Palette (Ctrl/Cmd+P) and run the same commands. The new note appears in the same folder with the correctly computed prefix.

Transformers dictate how the new note's name is generated. (Note: The transformers intentionally leave the text after the number blank so you can immediately type your new topic).

[!TIP] Have a unique naming convention? Please feel free to raise an issue or contribute a new transformer for your specific use case!

1. Increment Last Number

Finds the last number in your current note's title and increases it by 1. If no number is found, it appends a 1 to the end.

• 1 - Biology.md → 2 - .md
• 100 - Artificial Intelligence.md → 101 - .md

2. Nest Last Number

Adds a new sub-level number after the last decimal at the end of the note's index, nesting it one level deeper.

• 1 - Biology.md → 1.1 - .md
• 1.1 - Anatomy.md → 1.1.1 - .md

🔗 Related Notes

Manually hunting for connections between notes interrupts your thinking. Squire automatically surfaces related notes using lexical scoring — no tagging tax, no manual cross-referencing. Optionally, add semantic understanding by selecting a Transformer model in Settings.

Sidebar view: Opens a persistent panel that auto-refreshes as you switch notes. An algorithm badge in the header shows which scorer is active (TF‑IDF / Hybrid). Click a result to open it; click Link to insert [[Title]] at your cursor.

Quick modal: Ctrl/Cmd+P → Show related notes for a one-off look without changing your layout. Filter by typing.

Ribbon icon: The network icon toggles the sidebar view. The sidebar's open/closed state persists across Obsidian sessions.

Hybrid scoring: When a model is selected, Squire combines TF‑IDF and dense‑embedding cosine similarity into a single relevance score — so notes that use different words but share the same meaning still surface.

⚙️ Settings

Tune Squire to match your workflow without breaking your stride.

Index separator — the character(s) between your index number and title (e.g. - in 1 - Biology.md). Must be safe for file names, max 5 characters.

Number of suggestions — how many related notes to return (1–50, default 5).

Similarity scoring — choose TF‑IDF (lexical only) or select a Transformer model for hybrid semantic + lexical scoring. Models are downloaded on first use and cached in the plugin folder. Switch back to TF‑IDF any time by selecting the default.

Match weights — control how much word overlap, shared tags, and shared outgoing links influence similarity. Defaults: Words 1.0, Tags 0.5, Links 0.5. Set to 0 to disable a signal.

📦 Installation

1. Find Squire in Obsidian Community Plugins
2. Select "Add to Obsidian"

🤝 Contributing & Support

I accept Pull Requests, GitHub Issues, and general feedback! If you have a specific use case or transformer idea, please look at contributing to help make Squire better for everyone.

Funding: If you find this plugin useful and it saves you time, just let me know—hearing how it helps your workflow is greatly appreciated!