search_nodes
Query entities in structured data using Elasticsearch syntax. Filter by entity types, sort by relevance or recency, and specify zones for isolation. Enhances search precision with advanced operators like boolean, proximity, and wildcards for Knowledge Graph insights.
Instructions
Search entities using ElasticSearch query syntax. Supports boolean operators (AND, OR, NOT), fuzzy matching (~), phrases ("term"), proximity ("terms"~N), wildcards (*, ?), and boosting (^N). Examples: 'meeting AND notes', 'Jon~', '"project plan"~2'. All searches respect zone isolation.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| entityTypes | No | Filter to specific entity types (OR condition if multiple). | |
| includeObservations | No | Include full entity observations (default: false). | |
| informationNeeded | Yes | Important. Describe what information you are looking for, to give a precise context to the search engine AI agent. What questions are you trying to answer? Helps get more useful results. | |
| limit | No | Max results (default: 20, or 5 with observations). | |
| memory_zone | Yes | Limit search to specific zone. Omit for default zone. | |
| query | Yes | ElasticSearch query string. | |
| reason | Yes | Explain why this information is needed to help the AI agent give better results. The more context you provide, the better the results will be. | |
| sortBy | No | Sort by match quality, access time, or importance. |
Implementation Reference
- src/index.ts:326-370 (schema)JSON Schema defining the input parameters and validation for the 'search_nodes' tool, including query, informationNeeded, reason, entityTypes, limit, sortBy, includeObservations, and memory_zone.name: "search_nodes", description: "Search entities using ElasticSearch query syntax. Supports boolean operators (AND, OR, NOT), fuzzy matching (~), phrases (\"term\"), proximity (\"terms\"~N), wildcards (*, ?), and boosting (^N). Examples: 'meeting AND notes', 'Jon~', '\"project plan\"~2'. All searches respect zone isolation.", inputSchema: { type: "object", properties: { query: { type: "string", description: "ElasticSearch query string." }, informationNeeded: { type: "string", description: "Important. Describe what information you are looking for, to give a precise context to the search engine AI agent. What questions are you trying to answer? Helps get more useful results." }, reason: { type: "string", description: "Explain why this information is needed to help the AI agent give better results. The more context you provide, the better the results will be." }, entityTypes: { type: "array", items: {type: "string"}, description: "Filter to specific entity types (OR condition if multiple)." }, limit: { type: "integer", description: "Max results (default: 20, or 5 with observations)." }, sortBy: { type: "string", enum: ["relevance", "recency", "importance"], description: "Sort by match quality, access time, or importance." }, includeObservations: { type: "boolean", description: "Include full entity observations (default: false).", default: false }, memory_zone: { type: "string", description: "Limit search to specific zone. Omit for default zone." }, }, required: ["query", "memory_zone", "informationNeeded", "reason"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" }
- src/index.ts:979-996 (handler)MCP server tool dispatch handler for 'search_nodes' that extracts parameters from the request and delegates to KnowledgeGraphClient.userSearch for execution.else if (toolName === "search_nodes") { const includeObservations = params.includeObservations ?? false; const zone = params.memory_zone; // Use the high-level userSearch method that handles AI filtering internally const { entities: filteredEntities, relations: formattedRelations } = await kgClient.userSearch({ query: params.query, entityTypes: params.entityTypes, limit: params.limit, sortBy: params.sortBy, includeObservations, zone, informationNeeded: params.informationNeeded, reason: params.reason }); return formatResponse({ entities: filteredEntities, relations: formattedRelations }); }
- src/kg-client.ts:2328-2479 (handler)Core handler function in KnowledgeGraphClient that implements the search_nodes tool logic: performs Elasticsearch search with advanced query support, applies optional AI-based filtering using GroqAI based on informationNeeded and reason, adjusts entity relevance scores, retrieves related relations, and returns formatted results.async userSearch(params: { query: string; entityTypes?: string[]; limit?: number; includeObservations?: boolean; sortBy?: 'relevance' | 'recent' | 'importance'; zone?: string; informationNeeded?: string; reason?: string; }): Promise<{ entities: Array<{ name: string; entityType: string; observations?: string[]; lastRead?: string; lastWrite?: string; }>; relations: Array<{ from: string; to: string; type: string; fromZone: string; toZone: string; }>; }> { // Set default values const includeObservations = params.includeObservations ?? false; const defaultLimit = includeObservations ? 5 : 20; const zone = params.zone || this.defaultZone; const informationNeeded = params.informationNeeded; const reason = params.reason; // If informationNeeded is provided, increase the limit to get more results // that will be filtered later by the AI const searchLimit = informationNeeded ? (params.limit ? params.limit * 4 : defaultLimit * 4) : (params.limit || defaultLimit); // Prepare search parameters for the raw search const searchParams: ESSearchParams = { query: params.query, entityTypes: params.entityTypes, limit: searchLimit, sortBy: params.sortBy, includeObservations, zone, informationNeeded, reason }; // Perform the raw search const results = await this.search(searchParams); // Transform the results to a clean format, removing unnecessary fields const entities = results.hits.hits .filter(hit => hit._source.type === 'entity') .map(hit => { const entity: { name: string; entityType: string; observations?: string[]; lastRead?: string; lastWrite?: string; } = { name: (hit._source as ESEntity).name, entityType: (hit._source as ESEntity).entityType, }; // Only include observations and timestamps if requested if (includeObservations) { entity.observations = (hit._source as ESEntity).observations; entity.lastWrite = (hit._source as ESEntity).lastWrite; entity.lastRead = (hit._source as ESEntity).lastRead; } return entity; }); // Apply AI filtering if informationNeeded is provided and AI is available let filteredEntities = entities; if (informationNeeded && GroqAI.isEnabled && entities.length > 0) { try { // Get relevant entity names using AI filtering const usefulness = await GroqAI.filterSearchResults(entities, informationNeeded, reason); // If AI filtering returned null (error case), use original entities if (usefulness === null) { console.warn('AI filtering returned null, using original results'); filteredEntities = entities.slice(0, params.limit || defaultLimit); } else { // Filter entities to only include those with a usefulness score filteredEntities = entities.filter(entity => usefulness[entity.name] !== undefined ); // Sort entities by their relevance score from highest to lowest filteredEntities.sort((a, b) => { const scoreA = usefulness[a.name] || 0; const scoreB = usefulness[b.name] || 0; return scoreB - scoreA; }); const usefulEntities = filteredEntities.filter(entity => usefulness[entity.name] >= 60); const definatelyNotUsefulEntities = filteredEntities.filter(entity => usefulness[entity.name] < 20); // for each useful entities, increase the relevanceScore for (const entity of usefulEntities) { this.updateEntityRelevanceScore(entity.name, (usefulness[entity.name] + 45) * 0.01, zone); } // for each definately not useful entities, decrease the relevanceScore for (const entity of definatelyNotUsefulEntities) { this.updateEntityRelevanceScore(entity.name, 0.8 + usefulness[entity.name] * 0.01, zone); } // If no entities were found relevant, fall back to the original results if (filteredEntities.length === 0) { filteredEntities = entities.slice(0, params.limit || defaultLimit); } else { // Limit the filtered results to the requested amount filteredEntities = filteredEntities.slice(0, params.limit || defaultLimit); } } } catch (error) { console.error('Error applying AI filtering:', error); // Fall back to the original results but limit to the requested amount filteredEntities = entities.slice(0, params.limit || defaultLimit); } } else if (entities.length > (params.limit || defaultLimit)) { // If we're not using AI filtering but retrieved more results due to the doubled limit, // limit the results to the originally requested amount filteredEntities = entities.slice(0, params.limit || defaultLimit); } // Get relations between these entities const entityNames = filteredEntities.map(e => e.name); const { relations } = await this.getRelationsForEntities(entityNames, zone); // Map relations to a clean format const formattedRelations = relations.map(r => ({ from: r.from, to: r.to, type: r.relationType, fromZone: r.fromZone, toZone: r.toZone })); return { entities: filteredEntities, relations: formattedRelations }; }
- src/index.ts:89-649 (registration)Registration of all tools including 'search_nodes' via the ListToolsRequestSchema handler, which returns the complete list of available MCP tools with their schemas.server.setRequestHandler(ListToolsRequestSchema, async () => { return { tools: [ { name: "inspect_files", description: "Agent driven file inspection that uses AI to retrieve relevant content from multiple files.", inputSchema: { type: "object", properties: { file_paths: { type: "array", items: { type: "string" }, description: "Paths to the files (or directories) to inspect" }, information_needed: { type: "string", description: "Full description of what information is needed from the files, including the context of the information needed. Do not be vague, be specific. The AI agent does not have access to your context, only this \"information needed\" and \"reason\" fields. That's all it will use to decide that a line is relevant to the information needed. So provide a detailed specific description, listing all the details about what you are looking for." }, reason: { type: "string", description: "Explain why this information is needed to help the AI agent give better results. The more context you provide, the better the results will be." }, include_lines: { type: "boolean", description: "Whether to include the actual line content in the response, which uses more of your limited token quota, but gives more informatiom (default: false)" }, keywords: { type: "array", items: { type: "string" }, description: "Array of specific keywords related to the information needed. AI will target files that contain one of these keywords. REQUIRED and cannot be null or empty - the more keywords you provide, the better the results. Include variations, synonyms, and related terms." } }, required: ["file_paths", "information_needed", "include_lines", "keywords"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "inspect_knowledge_graph", description: "Agent driven knowledge graph inspection that uses AI to retrieve relevant entities and relations based on a query.", inputSchema: { type: "object", properties: { information_needed: { type: "string", description: "Full description of what information is needed from the knowledge graph, including the context of the information needed. Do not be vague, be specific. The AI agent does not have access to your context, only this \"information needed\" and \"reason\" fields. That's all it will use to decide that an entity is relevant to the information needed." }, reason: { type: "string", description: "Explain why this information is needed to help the AI agent give better results. The more context you provide, the better the results will be." }, include_entities: { type: "boolean", description: "Whether to include the full entity details in the response, which uses more of your limited token quota, but gives more information (default: false)" }, include_relations: { type: "boolean", description: "Whether to include the entity relations in the response (default: false)" }, keywords: { type: "array", items: { type: "string" }, description: "Array of specific keywords related to the information needed. AI will target entities that match one of these keywords. REQUIRED and cannot be null or empty - the more keywords you provide, the better the results. Include variations, synonyms, and related terms." }, memory_zone: { type: "string", description: "Memory zone to search in. If not provided, uses the default zone." }, entity_types: { type: "array", items: { type: "string" }, description: "Optional filter to specific entity types" } }, required: ["information_needed", "keywords"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "create_entities", description: "Create entities in knowledge graph (memory)", inputSchema: { type: "object", properties: { entities: { type: "array", items: { type: "object", properties: { name: {type: "string", description: "Entity name"}, entityType: {type: "string", description: "Entity type"}, observations: { type: "array", items: {type: "string"}, description: "Observations about this entity" } }, required: ["name", "entityType"] }, description: "List of entities to create" }, memory_zone: { type: "string", description: "Memory zone to create entities in." } }, required: ["entities", "memory_zone"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "update_entities", description: "Update entities in knowledge graph (memory)", inputSchema: { type: "object", properties: { entities: { type: "array", description: "List of entities to update", items: { type: "object", properties: { name: {type: "string"}, entityType: {type: "string"}, observations: { type: "array", items: {type: "string"} }, isImportant: {type: "boolean"} }, required: ["name"] } }, memory_zone: { type: "string", description: "Memory zone specifier. Entities will be updated in this zone." } }, required: ["entities", "memory_zone"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "delete_entities", description: "Delete entities from knowledge graph (memory)", inputSchema: { type: "object", properties: { names: { type: "array", items: {type: "string"}, description: "Names of entities to delete" }, memory_zone: { type: "string", description: "Memory zone specifier. Entities will be deleted from this zone." }, cascade_relations: { type: "boolean", description: "Whether to delete relations involving these entities (default: true)", default: true } }, required: ["names", "memory_zone"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "create_relations", description: "Create relationships between entities in knowledge graph (memory)", inputSchema: { type: "object", properties: { relations: { type: "array", description: "List of relations to create", items: { type: "object", properties: { from: {type: "string", description: "Source entity name"}, fromZone: {type: "string", description: "Optional zone for source entity, defaults to memory_zone or default zone. Must be one of the existing zones."}, to: {type: "string", description: "Target entity name"}, toZone: {type: "string", description: "Optional zone for target entity, defaults to memory_zone or default zone. Must be one of the existing zones."}, type: {type: "string", description: "Relationship type"} }, required: ["from", "to", "type"] } }, memory_zone: { type: "string", description: "Optional default memory zone specifier. Used if a relation doesn't specify fromZone or toZone." }, auto_create_missing_entities: { type: "boolean", description: "Whether to automatically create missing entities in the relations (default: true)", default: true } }, required: ["relations"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "delete_relations", description: "Delete relationships from knowledge graph (memory)", inputSchema: { type: "object", properties: { relations: { type: "array", description: "List of relations to delete", items: { type: "object", properties: { from: {type: "string", description: "Source entity name"}, to: {type: "string", description: "Target entity name"}, type: {type: "string", description: "Relationship type"} }, required: ["from", "to", "type"] } }, memory_zone: { type: "string", description: "Optional memory zone specifier. If provided, relations will be deleted from this zone." } }, required: ["relations"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "search_nodes", description: "Search entities using ElasticSearch query syntax. Supports boolean operators (AND, OR, NOT), fuzzy matching (~), phrases (\"term\"), proximity (\"terms\"~N), wildcards (*, ?), and boosting (^N). Examples: 'meeting AND notes', 'Jon~', '\"project plan\"~2'. All searches respect zone isolation.", inputSchema: { type: "object", properties: { query: { type: "string", description: "ElasticSearch query string." }, informationNeeded: { type: "string", description: "Important. Describe what information you are looking for, to give a precise context to the search engine AI agent. What questions are you trying to answer? Helps get more useful results." }, reason: { type: "string", description: "Explain why this information is needed to help the AI agent give better results. The more context you provide, the better the results will be." }, entityTypes: { type: "array", items: {type: "string"}, description: "Filter to specific entity types (OR condition if multiple)." }, limit: { type: "integer", description: "Max results (default: 20, or 5 with observations)." }, sortBy: { type: "string", enum: ["relevance", "recency", "importance"], description: "Sort by match quality, access time, or importance." }, includeObservations: { type: "boolean", description: "Include full entity observations (default: false).", default: false }, memory_zone: { type: "string", description: "Limit search to specific zone. Omit for default zone." }, }, required: ["query", "memory_zone", "informationNeeded", "reason"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "open_nodes", description: "Get details about specific entities in knowledge graph (memory) and their relations", inputSchema: { type: "object", properties: { names: { type: "array", items: {type: "string"}, description: "Names of entities to retrieve" }, memory_zone: { type: "string", description: "Optional memory zone to retrieve entities from. If not specified, uses the default zone." } }, required: ["names", "memory_zone"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "add_observations", description: "Add observations to an existing entity in knowledge graph (memory)", inputSchema: { type: "object", properties: { name: { type: "string", description: "Name of entity to add observations to" }, observations: { type: "array", items: {type: "string"}, description: "Observations to add to the entity" }, memory_zone: { type: "string", description: "Optional memory zone where the entity is stored. If not specified, uses the default zone." } }, required: ["memory_zone", "name", "observations"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "mark_important", description: "Mark entity as important in knowledge graph (memory) by boosting its relevance score", inputSchema: { type: "object", properties: { name: { type: "string", description: "Entity name" }, important: { type: "boolean", description: "Set as important (true - multiply relevance by 10) or not (false - divide relevance by 10)" }, memory_zone: { type: "string", description: "Optional memory zone specifier. If provided, entity will be marked in this zone." }, auto_create: { type: "boolean", description: "Whether to automatically create the entity if it doesn't exist (default: false)", default: false } }, required: ["memory_zone", "name", "important"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "get_recent", description: "Get recently accessed entities from knowledge graph (memory) and their relations", inputSchema: { type: "object", properties: { limit: { type: "integer", description: "Max results (default: 20 if includeObservations is false, 5 if true)" }, includeObservations: { type: "boolean", description: "Whether to include full entity observations in results (default: false)", default: false }, memory_zone: { type: "string", description: "Optional memory zone to get recent entities from. If not specified, uses the default zone." } }, required: ["memory_zone"], additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "list_zones", description: "List all available memory zones with metadata. When a reason is provided, zones will be filtered and prioritized based on relevance to your needs.", inputSchema: { type: "object", properties: { reason: { type: "string", description: "Reason for listing zones. What zones are you looking for? Why are you looking for them? The AI will use this to prioritize and filter relevant zones." } }, additionalProperties: false, "$schema": "http://json-schema.org/draft-07/schema#" } }, { name: "create_zone", description: "Create a new memory zone with optional description.", inputSchema: { type: "object", properties: { name: { type: "string", description: "Zone name (cannot be 'default')" }, shortDescription: { type: "string", description: "Short description of the zone." }, description: { type: "string", description: "Full zone description. Make it very descriptive and detailed." } }, required: ["name"] } }, { name: "delete_zone", description: "Delete a memory zone and all its entities/relations.", inputSchema: { type: "object", properties: { name: { type: "string", description: "Zone name to delete (cannot be 'default')" }, confirm: { type: "boolean", description: "Confirmation flag, must be true", default: false } }, required: ["name", "confirm"] } }, { name: "copy_entities", description: "Copy entities between zones with optional relation handling.", inputSchema: { type: "object", properties: { names: { type: "array", items: { type: "string" }, description: "Entity names to copy" }, source_zone: { type: "string", description: "Source zone" }, target_zone: { type: "string", description: "Target zone" }, copy_relations: { type: "boolean", description: "Copy related relationships (default: true)", default: true }, overwrite: { type: "boolean", description: "Overwrite if entity exists (default: false)", default: false } }, required: ["names", "source_zone", "target_zone"] } }, { name: "move_entities", description: "Move entities between zones (copy + delete from source).", inputSchema: { type: "object", properties: { names: { type: "array", items: { type: "string" }, description: "Entity names to move" }, source_zone: { type: "string", description: "Source zone" }, target_zone: { type: "string", description: "Target zone" }, move_relations: { type: "boolean", description: "Move related relationships (default: true)", default: true }, overwrite: { type: "boolean", description: "Overwrite if entity exists (default: false)", default: false } }, required: ["names", "source_zone", "target_zone"] } }, { name: "merge_zones", description: "Merge multiple zones with conflict resolution options.", inputSchema: { type: "object", properties: { source_zones: { type: "array", items: { type: "string" }, description: "Source zones to merge from" }, target_zone: { type: "string", description: "Target zone to merge into" }, delete_source_zones: { type: "boolean", description: "Delete source zones after merging", default: false }, overwrite_conflicts: { type: "string", enum: ["skip", "overwrite", "rename"], description: "How to handle name conflicts", default: "skip" } }, required: ["source_zones", "target_zone"] } }, { name: "zone_stats", description: "Get statistics for entities and relationships in a zone.", inputSchema: { type: "object", properties: { zone: { type: "string", description: "Zone name (omit for default zone)" } }, required: ["zone"] } }, { name: "get_time_utc", description: "Get the current UTC time in YYYY-MM-DD hh:mm:ss format", inputSchema: { type: "object", properties: {}, additionalProperties: false } } ] }; });
- legacy/index.ts:176-259 (helper)Legacy implementation of searchNodes in the file-based KnowledgeGraphManager (pre-Elasticsearch), provided for historical reference.async searchNodes(query: string): Promise<KnowledgeGraph> { const graph = await this.loadGraph(); // Get the basic search results with scores const searchResult = searchGraph(query, graph); // Create a map of entity name to search score for quick lookup // const searchScores = new Map<string, number>(); // searchResult.scoredEntities.forEach(scored => { // searchScores.set(scored.entity.name, scored.score); // }); // Find the maximum search score for normalization const maxSearchScore = searchResult.scoredEntities.length > 0 ? Math.max(...searchResult.scoredEntities.map(scored => scored.score)) : 1.0; // Get all entities sorted by lastRead date (most recent first) const entitiesByRecency = [...graph.entities] .filter(e => e.lastRead) // Filter out entities without lastRead .sort((a, b) => { // Sort in descending order (newest first) return new Date(b.lastRead!).getTime() - new Date(a.lastRead!).getTime(); }); // Get the 20 most recently accessed entities const top20Recent = new Set(entitiesByRecency.slice(0, 20).map(e => e.name)); // Get the 10 most recently accessed entities (subset of top20) const top10Recent = new Set(entitiesByRecency.slice(0, 10).map(e => e.name)); // Score the entities based on the criteria const scoredEntities = searchResult.scoredEntities.map(scoredEntity => { let score = 0; // Score based on recency if (top20Recent.has(scoredEntity.entity.name)) score += 1; if (top10Recent.has(scoredEntity.entity.name)) score += 1; // Score based on importance if (scoredEntity.entity.isImportant) { score += 1; score *= 2; // Double the score for important entities } // Add normalized search score (0-1 range) const searchScore = scoredEntity.score || 0; score += searchScore / maxSearchScore; return { entity: scoredEntity.entity, score }; }); // Sort by score (highest first) and take top 10 const topEntities = scoredEntities .sort((a, b) => b.score - a.score) .slice(0, 10) .map(item => item.entity); // Create a filtered graph with only the top entities const filteredEntityNames = new Set(topEntities.map(e => e.name)); const filteredRelations = graph.relations.filter(r => filteredEntityNames.has(r.from) || filteredEntityNames.has(r.to) ); const result: KnowledgeGraph = { entities: topEntities, relations: filteredRelations }; // Update access dates for found entities const todayFormatted = formatDate(); result.entities.forEach(foundEntity => { // Find the actual entity in the original graph and update its access date const originalEntity = graph.entities.find(e => e.name === foundEntity.name); if (originalEntity) { originalEntity.lastRead = todayFormatted; } }); // Save the updated access dates await this.saveGraph(graph); return result; }