MCP Notes
Enables persistent knowledge graph management with entities, observations, relations, temporal queries, and graph traversal via Model Context Protocol.
README
๐ง MCP Notes
MCP Notes is a powerful memory server that helps AI assistants remember and connect information over time. Think of it as a smart knowledge graph where you can store facts about people, projects, or concepts, link them together, and retrieve them intelligentlyโperfect for building AI agents with long-term memory.
Whether you're tracking user preferences, maintaining project context, or building relationship maps, MCP Notes provides a structured way to persist, search, and traverse your knowledge graph with confidence scores, timestamps, and rich metadata.
โจ Features
- ๐๏ธ Entity Management - Create, read, search, and delete entities with types and observations
- ๐ Observation Tracking - Add and remove observations with timestamps, confidence scores, and sources
- ๐ Relation Management - Create and delete relations between entities to build connections
- ๐ Knowledge Graph Operations - Read the entire graph, search for specific nodes, or traverse connected entities
- โฐ Temporal Queries - Query entities by time range using createdAt/updatedAt timestamps
- ๐ถ Graph Traversal - Walk relations to find connected entities (multi-hop BFS)
- ๐ Rich Metadata - Support for optional metadata, confidence scores, and source tracking
- ๐ Auto-Migration - Automatically migrates legacy plain-string observations to rich format on load
- ๐พ Persistent Storage - JSON file storage with daily logging for reliability
๐ Installation
# Clone the repository
git clone <your-repo-url>
cd mcp_notes
# Install dependencies
npm install
# Build the TypeScript code
npm run build
๐ฏ Quick Start
Running the Server
npm start
The server runs on stdio and connects to any MCP-compatible client.
Basic Example
{
"entities": [
{
"name": "Alice",
"entityType": "person",
"observations": [
"Alice is a software engineer",
{
"content": "Alice lives in San Francisco",
"confidence": 0.9,
"source": "user_profile"
}
]
}
]
}
Create relations between entities:
{
"relations": [
{
"from": "Alice",
"to": "Bob",
"relationType": "works with"
}
]
}
๐ ๏ธ Available Tools
create_entities
Create multiple new entities in the knowledge graph.
Parameters:
entities(array): Array of entity objects with:name(string): The name of the entityentityType(string): The type of the entityobservations(array): Initial observations (can be strings or objects withcontent,confidence,source)metadata(object, optional): Additional metadata for the entity
add_observations
Add new observations to existing entities.
Parameters:
observations(array): Array of observation additions with:entityName(string): Name of the entitycontents(array): Observations to add (can be strings or objects withcontent,confidence,source)
create_relations
Create relations between entities.
Parameters:
relations(array): Array of relation objects with:from(string): Source entity nameto(string): Target entity namerelationType(string): Type of relation (use active voice)
delete_entities
Delete entities and their associated relations.
Parameters:
entityNames(array of strings): Names of entities to delete
delete_observations
Delete specific observations from entities.
Parameters:
deletions(array): Array of deletion objects with:entityName(string): Name of the entityobservations(array of strings): Observations to delete
delete_relations
Delete relations from the knowledge graph.
Parameters:
relations(array): Array of relation objects to delete
open_nodes
Retrieve specific entities by name.
Parameters:
names(array of strings): Entity names to retrieve
read_graph
Read the entire knowledge graph including all entities and relations.
search_nodes
Search for entities matching a query.
Parameters:
query(string): Search query (matches entity names, types, and observation content)
query_by_time
Query entities and observations by time range using createdAt/updatedAt timestamps.
Parameters:
since(string, optional): ISO timestamp for start of time range (inclusive)until(string, optional): ISO timestamp for end of time range (inclusive)limit(number, optional): Maximum number of results to returnsort(string, optional): Sort order by updatedAt - "asc" or "desc" (default: "desc")
get_recent
Get the most recently updated entities.
Parameters:
limit(number, optional): Maximum number of recent entities to return (default: 10)
traverse_graph
Traverse the knowledge graph from a starting entity to find connected entities via relations.
Parameters:
start(string, required): Name of the entity to start traversal fromdepth(number, optional): Number of hops to traverse (default: 1)direction(string, optional): Direction of traversal - "out" (outgoing relations), "in" (incoming relations), or "both" (default: "both")
๐ ๏ธ Development
# Build
npm run build
# Watch mode for development
npm run dev
Type Definitions
TypeScript types are defined in src/types.ts for:
- Entity (with createdAt, updatedAt, metadata)
- Observation (with content, createdAt, confidence, source)
- Relation
- ObservationAddition
- ObservationDeletion
- RelationDeletion
- KnowledgeGraph
- GraphTraversalResult
๐ Data Model
Observations
Observations are rich objects with metadata:
interface Observation {
content: string;
createdAt: string; // ISO timestamp
confidence?: number; // 0-1
source?: string;
}
Entities
Entities include timestamps and optional metadata:
interface Entity {
name: string;
entityType: string;
observations: Observation[];
createdAt: string; // ISO timestamp
updatedAt: string; // ISO timestamp
metadata?: Record<string, unknown>;
}
Auto-Migration
Legacy data with plain-string observations is automatically migrated on load. Plain strings are converted to Observation objects with createdAt set to the current time.
โ๏ธ Configuration
Add this server to your MCP client configuration:
{
"mcpServers": {
"memory": {
"command": "node",
"args": ["/path/to/mcp_notes/dist/index.js"]
}
}
}
๐ Architecture
src/index.ts- Main MCP server with tool handlerssrc/memoryService.ts- Core memory service managing the knowledge graphsrc/types.ts- TypeScript type definitions
๐ License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.