Skip to content

Configuration

Vault Intelligence is designed to be powerful out of the box, but you can customise it to suit different hardware, budgets, and workflows.

Connection

SettingDefaultDescription
Google API keyNoneYour secret key from Google AI Studio. Stored in plain text in your plugin settings (data.json), but masked for display and logging using SHA-256 fingerprints. Required for all Gemini models and Gemini embeddings.
Refresh modelsNoneA manual trigger to force a fresh fetch of available models from the Gemini API.

Researcher

Consolidate settings for the Research Assistant agent.

SettingDefaultDescription
Chat modelgemini-3-flash-previewThe main intelligence engine.
Flash: Best for speed and agentic loops.
Pro: Best for deep reasoning or creative writing.
Agent languageEnglish (US)The primary language for all agent interactions. Affects the default system prompt and response style.
System instructionDefault (Managed)The core personality, role, and rules for the Researcher. Leave as default to receive automatic improvements in future updates.
Context window budget200,000Max tokens the AI can consider at once. Automatically scales proportionally when you switch models.
Max agent steps5Limits reasoning loops to prevent infinite "thinking" or high costs.
Web search modelgemini-2.5-flash-liteModel used specifically for web searches and fact-checking.
Enable computational solverOnAllows the agent to write and execute Python code for math and data analysis.
Code execution modelgemini-3-flash-previewThe specialised model used for generating Python code.
Vault reading limit25Max notes the Researcher can retrieve to answer a single question.

Explorer

Configure how connections and similar notes are discovered.

SettingDefaultDescription
Embedding providergeminiGoogle Gemini: Cloud-based. Requires API key.
Local: Offline. Runs on your CPU.
Embedding modelgemini-embedding-001The vector engine. Choose from Gemini presets or various local ONNX models.
Minimum similarity score0.5Relevance threshold (0.0 to 1.0). Matches below this are ignored.
Similar notes limit20Max number of related notes displayed in the sidebar.
GARS Similarity weight1.0Importance of vector similarity in hybrid scoring.
GARS Centrality weight0.2Importance of graph centrality (popularity) in hybrid scoring.
GARS Activation weight0.4Importance of semantic activation (connections) in hybrid scoring.
Re-index vaultNoneWipe and rebuild all embeddings. Required after changing models.

Gardener

Configure the Gardener agent for ontology maintenance and vault hygiene.

SettingDefaultDescription
Gardener modelgemini-3-flash-previewThe model used for analyzing vault structure and recommending improvements.
Gardener rulesDefault (Managed)The persona and hygiene instructions for the Gardener. Leave as default to receive automatic improvements.
Ontology pathOntologyFolder where concepts, entities, and MOCs are stored.
Gardener plans pathGardener/PlansFolder where proposed plans are saved.
Plans retention7 daysHow long to keep gardener plans before purging.
Excluded foldersDefaultFolders the gardener should ignore.
Recent note limit10Max notes to scan in a single session.
Context budget100,000Max token usage for a single gardener analysis.

Performance and System

Technical tuning for power users.

SettingDefaultDescription
Indexing delay5000msWait time after typing stops before re-indexing the current note.
Bulk scan delay300msDelay between files during full vault scans.
Local worker threads1-2CPU threads for local embeddings. Higher is faster but heavier.
Local SIMD accelerationAutoEnables SIMD instructions for local models. Faster but may be unstable on older hardware.
Gemini API retries10Number of retries for spotty connections.
Model cache duration7 daysDuration to cache Gemini model list locally.
Log levelWarnDeveloper console verbosity (Debug for full CoT).

Gemini vs Local Models

Google Gemini (Cloud)

  • Pros: Highest quality (gemini-embedding-001), zero local CPU/RAM overhead, handles large documents.
  • Cons: Requires API key, internet dependent, remote processing.

Transformers.js (Local)

  • Pros: 100% private, offline, no API costs.
  • Cons: Uses local resources, slightly lower quality on smaller presets.

Privacy and Git Sync

Vault Intelligence stores its search index and relationship graph in a specialized binary format inside the plugin's data/ directory.

  • Automated .gitignore: The plugin automatically creates and maintains a .gitignore file inside its internal data folder.
  • Why?: These index files can be very large (up to 100MB+ for massive vaults) and change frequently. Excluding them from Git prevents repository bloat and sync conflicts while using plugins like Obsidian Git.
  • Data safety: Since the index is a derived cache, it can be regenerated automatically from your notes at any time.