A
securityF
licenseA
qualityA Model Context Protocol (MCP) server that optimizes token usage by caching data during language model interactions, compatible with any language model and MCP client.
Last updated -
4
2
Implicitly supported through workspace isolation, allowing separate knowledge graphs for different Git projects
Referenced as the source of the original project that this MCP server was forked from
Supports installation and package management through npm for easy setup and dependency management
Mentioned in examples as an entity type that can be related to other entities in the knowledge graph
Provides time-based activity tracking through the get_temporal_events tool for historical data analysis
Utilizes TypeScript for type safety and improved developer experience when working with the MCP server API
Central Memory MCP Server
Model Context Protocol (MCP) memory server built with Azure Functions and TypeScript, providing persistent knowledge graph storage for AI assistants in VS Code.
Inspired by and forked from @modelcontextprotocol/server-memory
š Quick Start
npm install
func start
VS Code Integration
Install recommended extensions from
.vscode/extensions.jsonMCP configuration is ready in
.vscode/mcp.jsonUse in VS Code Copilot chat
Note: All MCP tools now use object parameters instead of JSON strings for better type safety and ease of use.
Test the Server
# Health check
curl http://localhost:7071/api/health
# Use in VS Code Copilot with object parameters:
# #memory-test_create_entities
# #memory-test_read_graph
# #memory-test_search_entities
Example Usage in VS Code Copilot
Recommended workflow for best results:
1. First, check existing data:
#memory-test_read_graph workspaceId="my-project"
2. Search for existing entities:
#memory-test_search_entities workspaceId="my-project" name="Alice"
3. Create entities (auto-updates existing ones):
#memory-test_create_entities workspaceId="my-project" entities={"name": "Alice", "entityType": "Person", "observations": ["Software engineer"]}
4. Create relationships (auto-creates missing entities):
#memory-test_create_relations workspaceId="my-project" relations={"from": "Alice", "to": "React Project", "relationType": "worksOn"}
5. Add observations (auto-creates entity if missing):
#memory-test_add_observation workspaceId="my-project" entityName="Alice" observation="Leads the frontend team" entityType="Person"
Key Features for Better LLM Usability:
ā Auto-creation of missing entities when adding observations or relations
ā Helpful error messages with examples when validation fails
ā Workflow guidance to view graph first, then search, then create
ā Clear parameter descriptions with expected formats
ā Reduced friction - tools handle common edge cases automatically
š§ MCP Tools
Core Operations:
read_graph- RECOMMENDED FIRST STEP: View the entire knowledge graph to understand existing datacreate_entities- Create entities with auto-update of existing onescreate_relations- Create relationships with auto-creation of missing entitiessearch_entities/search_relations- Search and verify existing dataadd_observation- Add observations with auto-creation of missing entitiesupdate_entity- Update entity observations and metadatadelete_entity- Remove entity and all its relationsget_stats- Get workspace statisticsclear_memory- Clear all workspace data
Recommended Workflow:
Use
read_graphto understand existing dataUse
search_entitiesto check for existing entitiesUse
create_entitiesto add new entitiesUse
create_relationsto connect entitiesUse
add_observationto add new information
Advanced Features:
get_temporal_events- Time-based activity trackingmerge_entities- Merge duplicate entitiesdetect_duplicate_entities- Find potential duplicatesexecute_batch_operations- Batch multiple operationsget_user_stats- Get user-specific statisticssearch_relations_by_user- Find relations by user
šļø Architecture
Built with:
Azure Functions v4 with TypeScript
Azure Table Storage for persistent data (via Azurite locally)
Model Context Protocol (MCP) for VS Code integration
Workspace isolation - each project gets separate storage
ļæ½ Project Structure
src/
āāā functions/ # Azure Functions endpoints
āāā services/ # Business logic (storage, entities, relations)
āāā types/ # TypeScript definitions
āāā index.ts # Main entry point
š Documentation
For detailed information, see the .docs/ folder:
Architecture Guide - Technical design and patterns
API Reference - Complete endpoint documentation
Storage Guide - Storage configuration and workspace management
Deployment Guide - Production deployment options
š Production Notes
Uses Azure Table Storage with managed identity for security
Workspace isolation prevents data leakage between projects
Health endpoints for monitoring and container orchestration
Automatic fallback to local storage for development
š License
MIT License - see LICENSE file for details.
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
Model Context Protocol (MCP) memory server built with Azure Functions and TypeScript, providing persistent knowledge graph storage for your preferred MCP client. Inspired by and forked from @modelcontextprotocol/server-memory
Related MCP Servers
- AsecurityFlicenseAqualityAn all-in-one Model Context Protocol (MCP) server that connects your coding AI to numerous databases, data warehouses, data pipelines, and cloud services, streamlining development workflow through seamless integrations.Last updated -3
- -securityAlicense-qualityModel Context Protocol (MCP) server that provides AI assistants with advanced web research capabilities, including Google search integration, intelligent content extraction, and multi-source synthesis.Last updated -114MIT License
- -securityFlicense-qualityA Model Context Protocol (MCP) server that provides persistent memory capabilities using OpenAI's vector stores, allowing AI assistants to save and search through memories across conversations.Last updated -2