MCP RAG Server POC
RenoLu/mcp-poc
Server MCP para RAG local — recuperação via cosine similarity, geração no cliente
Ver no GitHubServidor MCP minimalista que expõe a recuperação (retrieval) de um sistema RAG local via protocolo MCP. Integra-se com Ollama para embedding de queries e retorna chunks similares do vector store, deixando a geração de resposta para o cliente (Claude). Padrão pedagogicamente puro: separação clara entre recuperação e geração.
Instalação
npx convez-ai RenoLu/mcp-poc1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Integrar busca semantica em documentos locais com Claude Code/Desktop
- Demonstração didática de padrão RAG descentralizado (server retrieval, client generation)
- Prototipagem de sistemas que precisam citar fontes garantidas sem alucinação
Veredito

POC bem estruturado que exemplifica MCP corretamente — retrieval puro sem modelos de geração no server. Código educacional útil para quem quer entender padrão RAG modular ou construir agents que recuperam contexto determinístico antes de invocar Claude.
README
A tradução para o português deste README ainda está na fila.
mcp-poc
A small Model Context Protocol (MCP) server that exposes the retrieval step of the sibling rag-poc project as an MCP tool — "RAG over MCP."
MCP is a standard way to give an LLM client (Claude Code, Claude Desktop, …) access to tools that live outside the model. Here the pattern is deliberate: the server does retrieval only — it embeds your query, finds the most similar chunks in rag-poc's local vector store, and hands them back. The client's model does the generation, reading those chunks and writing a grounded, cited answer. The server never calls a chat model.
The tool
| Tool | Signature | What it does |
|---|---|---|
rag_search | (query: str, k: int = 4) -> list[dict] | Embeds query with the same local Ollama model that built the store, cosine-ranks the stored chunks, and returns the top k as {source, score, text} (most similar first). |
The model calling it is expected to answer from the returned chunks and cite each
source, or say it doesn't know if they don't contain the answer.
How it connects to rag-poc
This repo doesn't reimplement RAG — it imports rag-poc's rag package. The server
puts the rag-poc folder on sys.path and reuses its vector store, query embedder,
and input-sanitising hook. By default it expects rag-poc as a sibling folder
(../rag-poc); point elsewhere with the RAG_POC_PATH environment variable. The
store is read from RAG_POC_PATH/store.npz.
Prerequisites
- Ollama running at
localhost:11434with thenomic-embed-textmodel pulled (ollama pull nomic-embed-text). The query must be embedded by the same model that embedded the documents. - rag-poc ingested so its store exists — in the rag-poc folder:
python main.py ingest.
Setup
cd "C:\Coding Space\mcp-poc"
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
Test it — two ways
1. MCP Inspector (no Claude needed):
mcp dev server.py
Open the printed URL, pick rag_search, enter a query (e.g. "What is retrieval-augmented generation?"),
and inspect the returned chunks with their similarity scores.
2. From Claude Code. Register the server so rag_search appears in your session:
claude mcp add mcp-poc -- "C:\Coding Space\mcp-poc\.venv\Scripts\python.exe" "C:\Coding Space\mcp-poc\server.py"
If rag-poc is not a sibling of this repo, pass its location when registering:
claude mcp add mcp-poc --env RAG_POC_PATH="C:\path\to\rag-poc" -- "C:\Coding Space\mcp-poc\.venv\Scripts\python.exe" "C:\Coding Space\mcp-poc\server.py"
Then /mcp lists connected servers, and you can ask a question about your indexed
docs — Claude will call rag_search, pull the relevant chunks, and answer from them.
Remove it with claude mcp remove mcp-poc.
Next steps
- Add a
rag_answertool that runs rag-poc's full local pipeline (Ollama generation) to compare "the client model generates" vs "the local model generates." - Expose the indexed documents as MCP resources, or add a prompt template for a standard "answer with citations" instruction.