Skip to main content
Glama
mixelpixx

meMCP - Memory-Enhanced Model Context Protocol

memory_query

Query persistent memory to retrieve stored knowledge and context from previous LLM sessions, enabling continuous learning and information recall.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • Core handler function for 'memory_query' tool. Queries the factStore with provided args, formats results into a structured text response with scores, domains, tags etc., handles empty results and errors.
    async handleQuery(args) {
      try {
        const result = await this.factStore.queryFacts(args);
        
        if (result.facts.length === 0) {
          return {
            content: [
              {
                type: 'text',
                text: `No insights found matching your query: "${args.query}"`,
              },
            ],
          };
        }
        
        let response = `Found ${result.facts.length} insights (${result.total} total matches):\n\n`;
        
        for (const [index, fact] of result.facts.entries()) {
          response += `**${index + 1}. ${fact.type.replace('_', ' ')}** (Score: ${fact.qualityScore}/100)\n`;
          response += `*Domain: ${fact.domain}*\n`;
          response += `${fact.content}\n`;
          
          if (fact.tags && fact.tags.length > 0) {
            response += `*Tags: ${fact.tags.join(', ')}*\n`;
          }
          
          if (fact.relatedFacts && fact.relatedFacts.length > 0) {
            response += `*Related: ${fact.relatedFacts.length} connected facts*\n`;
          }
          
          if (fact.relevanceScore !== undefined) {
            response += `*Relevance: ${fact.relevanceScore.toFixed(2)}*\n`;
          }
          
          if (fact.semanticScore !== undefined) {
            response += `*Semantic: ${fact.semanticScore.toFixed(3)}*\n`;
          }
          
          response += `\n`;
        }
        
        return {
          content: [
            {
              type: 'text',
              text: response.trim(),
            },
          ],
        };
      } catch (error) {
        return {
          content: [
            {
              type: 'text',
              text: `Error querying insights: ${error.message}`,
            },
          ],
          isError: true,
        };
      }
    }
  • Input schema for the 'memory_query' tool defining parameters like query (required), filters for type, domain, tags, minScore, limit, includeRelated.
      type: 'object',
      properties: {
        query: {
          type: 'string',
          description: 'Search query to find relevant insights',
        },
        type: {
          type: 'string',
          description: 'Filter by fact type',
        },
        domain: {
          type: 'string',
          description: 'Filter by domain',
        },
        tags: {
          type: 'array',
          items: { type: 'string' },
          description: 'Filter by tags',
        },
        minScore: {
          type: 'number',
          description: 'Minimum quality score (0-100)',
          default: 0,
        },
        limit: {
          type: 'number',
          description: 'Maximum number of results',
          default: 10,
        },
        includeRelated: {
          type: 'boolean',
          description: 'Include related facts in results',
          default: false,
        },
      },
      required: ['query'],
    },
  • Direct registration of the 'memory_query' tool in registerQueryTool method, specifying name, description, schema, and handler.
    server.registerTool(
      'memory_query',
      'Search and retrieve relevant insights from the memory system',
      {
        type: 'object',
        properties: {
          query: {
            type: 'string',
            description: 'Search query to find relevant insights',
          },
          type: {
            type: 'string',
            description: 'Filter by fact type',
          },
          domain: {
            type: 'string',
            description: 'Filter by domain',
          },
          tags: {
            type: 'array',
            items: { type: 'string' },
            description: 'Filter by tags',
          },
          minScore: {
            type: 'number',
            description: 'Minimum quality score (0-100)',
            default: 0,
          },
          limit: {
            type: 'number',
            description: 'Maximum number of results',
            default: 10,
          },
          includeRelated: {
            type: 'boolean',
            description: 'Include related facts in results',
            default: false,
          },
        },
        required: ['query'],
      },
      async (args) => {
        return await this.handleQuery(args);
      }
    );
  • MemoryTools.registerTools method which calls queryHandler.registerTools(server), delegating the memory_query registration.
    async registerTools(server) {
      // Register tools from modular components
      this.operations.registerTools(server);
      this.queryHandler.registerTools(server);
      this.streamingTools.registerTools(server);
      this.management.registerTools(server);
    }
  • Top-level tool registration in SequentialGraphitiIntegration where MemoryTools (containing memory_query) is registered to the server.
    async registerTools(server) {
      if (!this.initialized) {
        throw new Error('Integration must be initialized before registering tools');
      }
    
      await this.memoryTools.registerTools(server);
      await this.configurationTools.registerTools(server);
    }

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/mixelpixx/meMCP'

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