Skip to main content
Glama
techRraj

MCP RAG + Agent Server for AI Tutor

by techRraj
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

MCP RAG + Agent Server for AI Tutor

A complete Model Context Protocol (MCP) server that combines:

  • RAG (Retrieval-Augmented Generation) with PostgreSQL + pgvector

  • Free LLM integration via OpenRouter

  • Dynamic agent creation as an MCP tool

  • FastMCP framework for building the server

Features

Tools

  • ingest_documents — Import documents into the vector knowledge base

  • search_knowledge — Semantic search using cosine similarity

  • ask_with_rag — Full RAG pipeline (retrieve + LLM generation)

  • create_agent — Dynamically create and run specialized AI agents

Resources

  • knowledge://stats — Knowledge base statistics

  • knowledge://config — Current server configuration

Prompts

  • explain_concept — Generate explanation prompts for any concept

Related MCP server: KB-MCP Server

Architecture

User → MCP Client (VS Code Copilot) → MCP Server → PostgreSQL+pgvector → OpenRouter LLM

Setup

1. Start PostgreSQL + pgvector

docker run -d --name pgvector-demo \
  -e POSTGRES_USER=tutor \
  -e POSTGRES_PASSWORD=tutor123 \
  -e POSTGRES_DB=ai_tutor \
  -p 5432:5432 pgvector/pgvector:pg16

##2. Install dependencies

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt

3. Configure .env

DATABASE_URL=postgresql://tutor:tutor123@localhost:5432/ai_tutor
OPENROUTER_API_KEY=sk-or-v1-your-key
OPENAI_API_KEY=sk-your-openai-key
OPENROUTER_MODEL=openrouter/free
EMBEDDING_MODEL=nvidia/nemotron-3-embed-1b:free
EMBEDDING_BASE_URL=https://openrouter.ai/api/v1



4. Run the server

python server.py


Testing
Use MCP Inspector:
fastmcp dev inspector server.py

Tech Stack
FastMCP (MCP SDK for Python)

PostgreSQL 16 + pgvector

OpenAI Python SDK (via OpenRouter)

Python 3.11+

License
Apache 2.0

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

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • RAG-MCP Knowledge Base Server

    Luxter77
    A
    license
    Not graded
    quality
    D
    maintenance
    Enables semantic search across text documents using vector embeddings stored in PostgreSQL. Provides multiple search modalities including semantic similarity, question/answer, and style-based search through a retrieval-augmented generation system.
    3
    Apache 2.0

  • KB-MCP Server

    isshiki-dev
    F
    license
    A
    quality
    Not graded
    maintenance
    A local-first knowledge base server that enables AI clients to store, retrieve, and manage documents using semantic search. Provides privacy-focused, offline-capable memory for AI assistants with tools for ingesting, querying, updating, and deleting knowledge.
    7
    191 npm
    -

  • rag-mcp

    NextLevelManagementAdvisors
    A
    license
    A
    quality
    B
    maintenance
    Enables ingestion and semantic search over text documents using PostgreSQL + pgvector and OpenAI-compatible embeddings, allowing any LLM agent to retrieve relevant chunks for grounded answers.
    4
    AGPL 3.0