Which integrations are available for this server?

Uses LangChain for LLM orchestration and chaining within the RAG pipeline to process queries and generate responses. Implements a state machine architecture for the RAG pipeline with multi-stage processing including guardrails, Cypher generation, retrieval, and response generation. Provides a RAG system that converts natural language queries into Cypher queries to retrieve information from a Neo4j graph database, with support for dynamic graph schema configuration and intelligent query routing. Integrates OpenAI models for dual LLM strategy: fast models for guardrails decision-making and accurate models for Cypher query generation and answering.

How do I use SOLVRO MCP - Knowledge Graph RAG System?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SOLVRO MCP - Knowledge Graph RAG System What courses are taught by Professor Kowalski in the Computer Science department?" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

de en es ja ko ru zh

SOLVRO MCP - Knowledge Graph RAG System

Official

by Solvro

Overview Schema Source Related Servers Discussions Admin

Python

Hybrid

 backend-mcp (separate repo)   │            ml-mcp (this repo, no host ports)
┌─────────────┐   ┌──────────────┐│   ┌─────────────┐          ┌─────────────┐
│    nginx    │──▶│ chat-service │┼──▶│  MCP Server │─────────▶│    Neo4j    │
│   :8080     │   │              ││   │ :8005 (int.)│          │ :7687 (int.)│
└─────────────┘   └──────────────┘│   └─────────────┘          └─────────────┘
   auth + UI          agent       │ solvro-mcp-internal          mcp_network

Intelligent Query Routing - Guardrails system determines query relevance
Natural Language to Cypher - Converts questions to graph queries
Knowledge Graph RAG - Retrieval-Augmented Generation with Neo4j
MCP Protocol - Standard Model Context Protocol interface
Observability - Optional Langfuse tracing integration
Docker Ready - One command deployment

Quick Start

# Setup
just setup
cp .env.example .env  # Edit with your API keys

# Run with Docker
just up      # Neo4j + MCP Server, reachable only by backend-mcp (no host ports)
just up-dev  # same, plus 127.0.0.1 ports for local work
just logs    # View logs
just down    # Stop services

Related MCP server: Unreal Engine Knowledge Graph MCP Server

Architecture

System Overview

 backend-mcp (separate repo)   │            ml-mcp (this repo, no host ports)
┌─────────────┐   ┌──────────────┐│   ┌─────────────┐          ┌─────────────┐
│    nginx    │──▶│ chat-service │┼──▶│  MCP Server │─────────▶│    Neo4j    │
│   :8080     │   │              ││   │ :8005 (int.)│          │ :7687 (int.)│
└─────────────┘   └──────────────┘│   └─────────────┘          └─────────────┘
   auth + UI          agent       │ solvro-mcp-internal          mcp_network

Service	Container port	Reachable from	Description
`mcp-server`	8005	`backend-mcp` over `solvro-mcp-internal`	FastMCP server exposing `knowledge_graph_tool` and `/health`
`neo4j`	7474/7687	`mcp-server` over `mcp_network` only	Knowledge graph database

The chat UI and the HTTP API that users talk to live in backend-mcp; this repository is the graph, the retrieval pipeline and the ETL that fills it.

Nothing is published on the host. just up-dev layers docker/compose.dev.yml on top, which republishes the ports on 127.0.0.1 for the Neo4j browser, just kg and uv run dump-graph.

RAG Pipeline

The heart of the system is a LangGraph-based RAG pipeline that intelligently processes user queries:

Pipeline Flow:

Guardrails - Fast LLM determines if query is relevant to knowledge base
Cypher Generation - Accurate LLM converts natural language to Cypher query
Retrieval - Execute query against Neo4j knowledge graph
Response - Return structured context data

Data Pipeline

Separate ETL pipeline for ingesting documents into the knowledge graph:

Pipeline Steps:

Document Loading - PDF and text document ingestion
Text Extraction - OCR and content extraction
LLM Processing - Generate Cypher queries from content
Graph Population - Execute queries to build knowledge graph

Configuration

Copy .env.example to .env and configure:

########################################
# LLM / AI Provider Keys
########################################

# OpenAI API key (optional)
OPENAI_API_KEY=

# DeepSeek API key (optional)
DEEPSEEK_API_KEY=

# Google Generative AI / PaLM API key (optional)
GOOGLE_API_KEY=

# CLARIN LLM API key (optional, used by API & client)
CLARIN_API_KEY=


########################################
# Logging
########################################

# Root log level for every entry point: DEBUG, INFO, WARNING, ERROR or CRITICAL
LOG_LEVEL=INFO


########################################
# Langfuse Observability
########################################

LANGFUSE_SECRET_KEY=
LANGFUSE_PUBLIC_KEY=
LANGFUSE_HOST=https://cloud.langfuse.com


########################################
# Neo4j Database
########################################

# URI used by data pipeline, MCP server and graph config
NEO4J_URI=bolt://localhost:7687
NEO4J_USER=neo4j
NEO4J_PASSWORD=


########################################
# Data Pipeline Runtime Controls
########################################

# Max parallel pages processed per batch
DATA_PIPELINE_MAX_CONCURRENCY=4

# Minutes after which a stuck in-progress hash can be reclaimed
DATA_PIPELINE_CLAIM_STALE_MINUTES=30


########################################
# MCP Server Networking
########################################

# Bind host for the MCP server process
MCP_BIND_HOST=0.0.0.0

# Host/port used by API and MCP client to reach the MCP server
MCP_HOST=127.0.0.1
MCP_PORT=8005

Commands

# Docker Stack
just up          # Neo4j + MCP server, no host ports
just up-dev      # same, plus 127.0.0.1 ports for local work
just down        # Stop services
just logs        # View logs
just ps          # Service status
just nuke        # Remove everything

# Local Development
just mcp-server  # Run MCP server
just kg "query"  # Query knowledge graph

# Quality
just lint        # Format & lint
just test        # Run tests
just ci          # Full CI pipeline
uv run --with pytest python -m pytest tests/data_pipeline/test_pipeline_concurrency.py -q
                # Run pipeline concurrency/idempotency tests only

# Data Pipeline
just prefect-up  # Start Prefect (UI on 127.0.0.1:4200 only)
just pipeline    # Run ETL

Project Structure

src/
├── mcp_server/      # MCP server + RAG pipeline
├── mcp_client/      # CLI client
├── config/          # Configuration
└── data_pipeline/   # Prefect ETL flows

docker/
├── compose.stack.yml    # Main stack (Neo4j + MCP server, no host ports)
├── compose.dev.yml      # Override that republishes the ports on 127.0.0.1
├── compose.prefect.yml  # Data pipeline
├── Dockerfile.mcp       # MCP server image
└── Dockerfile.prefect   # Data pipeline image

Querying the Server

The server speaks MCP over HTTP at http://mcp-server:8005/mcp on the shared network. From the host, bring the stack up with just up-dev and use the CLI:

just kg "Czym jest nagroda dziekana?"

GET http://127.0.0.1:8005/health answers 200 {"status": "healthy"} once the server can reach Neo4j, and 503 with a reason otherwise. The user-facing chat endpoint, sessions and authentication are in backend-mcp.

Tech Stack

Technology	Purpose
FastMCP	Model Context Protocol server
LangGraph	RAG state machine
LangChain	LLM orchestration
Neo4j	Knowledge graph database
Langfuse	Observability (optional)
Prefect	Data pipeline orchestration
Docker	Containerization

License

MIT © Solvro

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityActive

ResponsivenessWithin a week

Resources

GitHub Repository

Need Help?

Related Servers

Related MCP Connectors

DocsMint A
io.github.HiAi-gg
Cloud or self-hosted knowledge for AI agents: hybrid search, reranking, GraphRAG, scoped MCP tools.
Documentation Access Knowledge & Memory RAG Systems
21171
BrainstormRouter MCP Server
io.github.justinjilg
AI routing, memory, guardrails, and governance. Routes across Claude, GPT, Gemini.
RAG Systems Knowledge & Memory Agent Orchestration
2
Konstruct
io.github.alright-alright
Knowledge graph for AI agents. Query concepts, walk edges, get advisories.
Knowledge & Memory RAG Systems Databases
2
ragie
com.mcparmory
Ingest, manage, and retrieve documents for RAG-powered AI applications
RAG Systems Databases Knowledge & Memory
2

Related MCP Servers

Neo4j Knowledge Graph
Knowledge & Memory Databases
mjftw
F
license
Not graded
quality
D
maintenance
Enables storage and retrieval of knowledge in a graph database format, allowing users to create, update, search, and delete entities and relationships in a Neo4j-powered knowledge graph through natural language.
Updated a year ago (2025-03-30 17:41 UTC)
5
-
Unreal Engine Knowledge Graph
Knowledge & Memory Documentation Access Developer Tools
yarnovo
A
license
Not graded
quality
C
maintenance
Enables intelligent exploration of Unreal Engine documentation through a Neo4j-based knowledge graph. Supports concept relationship searches, learning path discovery, and dependency tracking between engine features using bilingual Chinese-English queries.
Updated a year ago (2025-06-16 11:11 UTC)
4 npm
7
ISC
Neo4j MCP Server
Databases Knowledge & Memory RAG Systems
vpro1032
A
license
Not graded
quality
D
maintenance
Enables AI assistants to interact with Neo4j graph databases through natural language, supporting Cypher queries, schema management, data manipulation, and graph algorithms.
Updated 9 months ago (2025-12-07 22:33 UTC)
MIT
Football Knowledge Graph RAG
RAG Systems Knowledge & Memory Sports
Fachreza28
F
license
Not graded
quality
D
maintenance
Enables natural language querying of a football knowledge graph via Neo4j, generating Cypher queries and answers using LLMs, and also provides tools to build the graph from text.
Updated 3 months ago (2026-06-23 07:11 UTC)
-