delorenj/mcp-qdrant-memory

This MCP server provides a knowledge graph implementation with semantic search capabilities powered by Qdrant vector database.
## Features
Graph-based knowledge representation with entities and relations
File-based persistence (memory.json)
Semantic search using Qdrant vector database
OpenAI embeddings for semantic similarity
HTTPS support with reverse proxy compatibility
Docker support for easy deployment
## Environment Variables
The following environment variables are required:
bash
# OpenAI API key for generating embeddings
OPENAIAPIKEY=your-openai-api-key
# Qdrant server URL (supports both HTTP and HTTPS)
QDRANTURL=https://your-qdrant-server
# Qdrant API key (if authentication is enabled)
QDRANTAPIKEY=your-qdrant-api-key
# Name of the Qdrant collection to use
QDRANTCOLLECTIONNAME=your-collection-name
## Setup
### Local Setup
1. Install dependencies:
bash
npm install
2. Build the server:
bash
npm run build
### Docker Setup
1. Build the Docker image:
bash
docker build -t mcp-qdrant-memory .
2. Run the Docker container with required environment variables:
bash
docker run -d \
-e OPENAIAPIKEY=your-openai-api-key \
-e QDRANTURL=http://your-qdrant-server:6333 \
-e QDRANTCOLLECTIONNAME=your-collection-name \
-e QDRANTAPIKEY=your-qdrant-api-key \
--name mcp-qdrant-memory \
mcp-qdrant-memory
### Add to MCP settings:
json
"mcpServers": {
"memory": {
"command": "/bin/zsh",
"args": ["-c", "cd /path/to/server && node dist/index.js"],
"env": {
"OPENAIAPIKEY": "your-openai-api-key",
"QDRANTAPIKEY": "your-qdrant-api-key",
"QDRANTURL": "http://your-qdrant-server:6333",
"QDRANTCOLLECTIONNAME": "your-collection-name"
},
"alwaysAllow": [
"createentities",
"createrelations",
"addobservations",
"deleteentities",
"deleteobservations",
"deleterelations",
"readgraph",
"searchsimilar"
## Tools
### Entity Management
createentities: Create multiple new entities
createrelations: Create relations between entities
addobservations: Add observations to entities
deleteentities: Delete entities and their relations
deleteobservations: Delete specific observations
deleterelations: Delete specific relations
readgraph: Get the full knowledge graph
### Semantic Search
searchsimilar: Search for semantically similar entities and relations
typescript
interface SearchParams {
query: string; // Search query text
limit?: number; // Max results (default: 10)
## Implementation Details
The server maintains two forms of persistence:
1. File-based (memory.json):
Complete knowledge graph structure
Fast access to full graph
Used for graph operations
2. Qdrant Vector DB:
Semantic embeddings of entities and relations
Enables similarity search
Automatically synchronized with file storage
### Synchronization
When entities or relations are modified:
1. Changes are written to memory.json
2. Embeddings are generated using OpenAI
3. Vectors are stored in Qdrant
4. Both storage systems remain consistent
### Search Process
When searching:
1. Query text is converted to embedding
2. Qdrant performs similarity search
3. Results include both entities and relations