Skip to main content
Glama
msdanyg

Smart Connections MCP Server

by msdanyg
index.js12 kB
#!/usr/bin/env node /** * Smart Connections MCP Server * * Provides semantic search and knowledge graph capabilities for Obsidian Smart Connections * via the Model Context Protocol (MCP). */ import { Server } from '@modelcontextprotocol/sdk/server/index.js'; import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js'; import { CallToolRequestSchema, ListToolsRequestSchema, } from '@modelcontextprotocol/sdk/types.js'; import { z } from 'zod'; import { SmartConnectionsLoader } from './smart-connections-loader.js'; import { SearchEngine } from './search-engine.js'; // Environment variable for vault path const VAULT_PATH = process.env.SMART_VAULT_PATH; if (!VAULT_PATH) { console.error('Error: SMART_VAULT_PATH environment variable is required'); console.error('Please set it to your Obsidian vault path, e.g.:'); console.error(' export SMART_VAULT_PATH="/Users/username/My Vault"'); process.exit(1); } // Initialize loader const loader = new SmartConnectionsLoader(VAULT_PATH); await loader.initialize(); // Create search engine after loader is initialized const searchEngine = new SearchEngine(loader); console.error('Smart Connections MCP Server initialized successfully'); console.error(`Vault: ${VAULT_PATH}`); console.error(`Loaded ${loader.getSources().size} notes`); // Create MCP server const server = new Server({ name: 'smart-connections-mcp', version: '1.0.0', }, { capabilities: { tools: {}, }, }); // Define tool schemas const GetSimilarNotesSchema = z.object({ note_path: z.string().describe('Path to the note (e.g., "Note.md" or "Folder/Note.md")'), threshold: z.number().min(0).max(1).default(0.5).describe('Similarity threshold (0-1)'), limit: z.number().int().positive().default(10).describe('Maximum number of results'), }); const GetConnectionGraphSchema = z.object({ note_path: z.string().describe('Path to the note to start from'), depth: z.number().int().positive().default(2).describe('Depth of the connection graph'), threshold: z.number().min(0).max(1).default(0.6).describe('Similarity threshold (0-1)'), max_per_level: z.number().int().positive().default(5).describe('Max connections per level'), }); const SearchNotesSchema = z.object({ query: z.string().describe('Search query text'), limit: z.number().int().positive().default(10).describe('Maximum number of results'), threshold: z.number().min(0).max(1).default(0.5).describe('Similarity threshold (0-1)'), }); const GetEmbeddingNeighborsSchema = z.object({ embedding_vector: z.array(z.number()).describe('384-dimensional embedding vector'), k: z.number().int().positive().default(10).describe('Number of neighbors to return'), threshold: z.number().min(0).max(1).default(0.5).describe('Similarity threshold (0-1)'), }); const GetNoteContentSchema = z.object({ note_path: z.string().describe('Path to the note'), include_blocks: z.array(z.string()).optional().describe('Specific block headings to include'), }); const GetStatsSchema = z.object({}); // Define available tools const tools = [ { name: 'get_similar_notes', description: 'Find notes semantically similar to a given note using embeddings. Returns paths, similarity scores, and available blocks.', inputSchema: { type: 'object', properties: { note_path: { type: 'string', description: 'Path to the note (e.g., "Note.md" or "Folder/Note.md")', }, threshold: { type: 'number', description: 'Similarity threshold (0-1), default 0.5', minimum: 0, maximum: 1, default: 0.5, }, limit: { type: 'number', description: 'Maximum number of results, default 10', minimum: 1, default: 10, }, }, required: ['note_path'], }, }, { name: 'get_connection_graph', description: 'Build a multi-level connection graph starting from a note, showing how notes are semantically connected.', inputSchema: { type: 'object', properties: { note_path: { type: 'string', description: 'Path to the note to start from', }, depth: { type: 'number', description: 'Depth of the connection graph (levels), default 2', minimum: 1, default: 2, }, threshold: { type: 'number', description: 'Similarity threshold (0-1), default 0.6', minimum: 0, maximum: 1, default: 0.6, }, max_per_level: { type: 'number', description: 'Max connections per level, default 5', minimum: 1, default: 5, }, }, required: ['note_path'], }, }, { name: 'search_notes', description: 'Search for notes using a text query. Returns notes ranked by relevance with similarity scores.', inputSchema: { type: 'object', properties: { query: { type: 'string', description: 'Search query text', }, limit: { type: 'number', description: 'Maximum number of results, default 10', minimum: 1, default: 10, }, threshold: { type: 'number', description: 'Similarity threshold (0-1), default 0.5', minimum: 0, maximum: 1, default: 0.5, }, }, required: ['query'], }, }, { name: 'get_embedding_neighbors', description: 'Find nearest neighbors for a given embedding vector. Useful for custom similarity searches.', inputSchema: { type: 'object', properties: { embedding_vector: { type: 'array', items: { type: 'number' }, description: '384-dimensional embedding vector', }, k: { type: 'number', description: 'Number of neighbors to return, default 10', minimum: 1, default: 10, }, threshold: { type: 'number', description: 'Similarity threshold (0-1), default 0.5', minimum: 0, maximum: 1, default: 0.5, }, }, required: ['embedding_vector'], }, }, { name: 'get_note_content', description: 'Retrieve the full content of a note, optionally with specific blocks/sections extracted.', inputSchema: { type: 'object', properties: { note_path: { type: 'string', description: 'Path to the note', }, include_blocks: { type: 'array', items: { type: 'string' }, description: 'Specific block headings to include (optional)', }, }, required: ['note_path'], }, }, { name: 'get_stats', description: 'Get statistics about the Smart Connections knowledge base (total notes, blocks, embedding model, etc.).', inputSchema: { type: 'object', properties: {}, }, }, ]; // Handle tool list requests server.setRequestHandler(ListToolsRequestSchema, async () => { return { tools }; }); // Handle tool execution requests server.setRequestHandler(CallToolRequestSchema, async (request) => { const { name, arguments: args } = request.params; try { switch (name) { case 'get_similar_notes': { const { note_path, threshold, limit } = GetSimilarNotesSchema.parse(args); const results = searchEngine.getSimilarNotes(note_path, threshold, limit); return { content: [ { type: 'text', text: JSON.stringify(results, null, 2), }, ], }; } case 'get_connection_graph': { const { note_path, depth, threshold, max_per_level } = GetConnectionGraphSchema.parse(args); const graph = searchEngine.getConnectionGraph(note_path, depth, threshold, max_per_level); return { content: [ { type: 'text', text: JSON.stringify(graph, null, 2), }, ], }; } case 'search_notes': { const { query, limit, threshold } = SearchNotesSchema.parse(args); const results = searchEngine.searchByQuery(query, limit, threshold); return { content: [ { type: 'text', text: JSON.stringify(results, null, 2), }, ], }; } case 'get_embedding_neighbors': { const { embedding_vector, k, threshold } = GetEmbeddingNeighborsSchema.parse(args); const results = searchEngine.getEmbeddingNeighbors(embedding_vector, k, threshold); return { content: [ { type: 'text', text: JSON.stringify(results, null, 2), }, ], }; } case 'get_note_content': { const { note_path, include_blocks } = GetNoteContentSchema.parse(args); const result = searchEngine.getNoteWithContext(note_path, include_blocks); return { content: [ { type: 'text', text: JSON.stringify(result, null, 2), }, ], }; } case 'get_stats': { GetStatsSchema.parse(args); const stats = searchEngine.getStats(); return { content: [ { type: 'text', text: JSON.stringify(stats, null, 2), }, ], }; } default: throw new Error(`Unknown tool: ${name}`); } } catch (error) { const errorMessage = error instanceof Error ? error.message : String(error); return { content: [ { type: 'text', text: JSON.stringify({ error: errorMessage }, null, 2), }, ], isError: true, }; } }); // Start the server const transport = new StdioServerTransport(); await server.connect(transport); console.error('Smart Connections MCP Server running on stdio'); //# sourceMappingURL=index.js.map

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/msdanyg/smart-connections-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server