Skip to content

Settings

Jump to bottom
dhwang0803-glitch edited this page Jul 24, 2026 · 2 revisions

Settings Reference

Access via Settings → Community Plugins → Vaultend.

Language

Setting Options Default
Display Language Auto (follow Obsidian) / English / Korean Auto

Views update immediately when changed. Command palette names update after restarting Obsidian.

AI Provider

Setting Description Default
AI Provider Select the AI service: OpenAI, Google Gemini, Ollama (Local), or Custom (OpenAI-compatible) OpenAI
API Key Your provider's API key (not needed for Ollama) —
Model Select from the dropdown or choose "Custom" to enter any model ID gpt-4o
Embedding Model Model for semantic similarity (tag matching, link suggestions, duplicate detection) Auto-detected per provider

Changes apply immediately — no restart needed.

Provider-Specific Settings

Ollama (Local):

Setting Description Default
Ollama Base URL URL of the Ollama server http://localhost:11434

Ollama runs models locally — no API key needed. Install models with ollama pull <model>. For embedding features, also install an embedding model (e.g., nomic-embed-text).

Custom (OpenAI-compatible):

Setting Description Default
Custom Base URL Base URL for the API endpoint —
Custom API Key API key for the endpoint (optional) —
Custom Model Model name to use —

Embedding Support by Provider

Provider Chat/Completion Embeddings
OpenAI Yes Yes
Google Gemini Yes Yes
Ollama (Local) Yes Yes (requires embedding model)
Custom Yes Depends on endpoint

Embedding-based features include: tag similarity ranking, cross-language tag resolution, and link suggestions. These features are degraded without embedding support.

Default Embedding Models

Provider Default Embedding Model
OpenAI text-embedding-3-small
Google Gemini gemini-embedding-001
Ollama nomic-embed-text

Token & Cost Display

Organize modals and folder results show token usage and estimated cost after each AI call.

Item Accuracy Source
Token count Exact Returned by the AI provider's API response
Estimated cost Approximate Calculated from a built-in pricing table

Why cost is approximate:

  • The pricing table is updated with each plugin release, but may lag behind provider price changes
  • Cached-token discounts, batch discounts, and free-tier credits are not reflected
  • Models not in the table show "Cost: N/A" instead of an incorrect estimate
  • Ollama (local) always shows $0.00 (no API cost)

For exact billing, check your provider's usage dashboard:

Organize Folder

Setting Description Default
Auto-apply results When enabled, AI results are applied immediately without manual review Off
Max Response Tokens Maximum tokens for AI response (1024–16384). Higher values cost more. 8192

Link Suggestion

Setting Description Default
Similarity Threshold Minimum cosine similarity for embedding-based link suggestions (0.30–0.80). Lower values suggest more links but may include less relevant ones. 0.40

Maintenance

Setting Description Default
Auto Maintenance Run vault maintenance periodically Off
Maintenance Interval Minutes between automatic scans 60
Reject Suppression (days) Days to suppress a rejected proposal before showing it again 7
Exclude Folders Folders excluded from maintenance scans (chip UI with vault autocomplete) —
Exclude Tags Notes with these tags are skipped during scans (chip UI with vault autocomplete) —
Archive Folder Destination folder for archived notes Archive

Maintenance scans only Markdown (.md) notes. Non-text files (Excalidraw, Canvas) are automatically excluded.

Search (Advanced)

Setting Description Default
Embedding weight Weight multiplier for embedding results in hybrid search. Higher values favor semantic similarity over keyword match. 4.0
RRF K parameter Reciprocal Rank Fusion smoothing parameter. Lower values make rank differences more dramatic. 20

Privacy

See Privacy Rules for detailed configuration guide.

Setting Description
Privacy Rules List of rules controlling what data is sent to AI. Each rule has a name, type, pattern, and enable/disable toggle.

Internal Data

Vaultend stores internal data in .vaultend/ inside your vault:

File Purpose
search-index.json BM25 keyword search index
embeddings.json Vector store for note embeddings
tag-embeddings.json Tag embedding cache
note-embeddings.json Note embedding cache
tag-groups.json Tag group cache (Organize Tags)
tfidf-corpus.json TF-IDF corpus stats
dirty-set.json Changed files tracker (smart scheduling)
history/ Activity log entries

This folder is automatically created and managed by the plugin. You generally don't need to interact with it directly.

Note: If you move or sync your vault, the .vaultend/ folder should be included. Embedding caches may need to regenerate if the AI provider changes.

Vaultend · Issues · Releases · MIT License

Vaultend Wiki

Features

Reference

Clone this wiki locally