convez.ai
MCP ServerDados & AnalyticsIntermediário
Avishkar-Kabadi-ODA

LibreChat Analytics MCP Server

Avishkar-Kabadi-ODA/librechat-analytics-mcp

Analytics em tempo real do LibreChat via MCP Server

Ver no GitHub
InicialPluginScore · de 0 a 100

Servidor MCP que conecta diretamente ao banco de dados MongoDB do LibreChat para fornecer análises em tempo real sobre consumo de tokens, uso de modelos e atividade de usuários. Oferece ferramentas para consultar métricas agregadas de uso, identificar top users e acompanhar tendências de consumo.

Instalação

$npx convez-ai Avishkar-Kabadi-ODA/librechat-analytics-mcp

1 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

Estrelas
0
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
1

Casos de uso

  • Monitorar consumo de tokens por modelo em instâncias LibreChat
  • Identificar usuários com maior consumo de recursos
  • Gerar relatórios de uso agregados (diário, mensal, trimestral, anual)
  • Rastrear atividade de usuários e mensagens processadas

Veredito

Ferramenta especializada e bem documentada para quem opera instâncias LibreChat em produção. Resolve problema real de observabilidade com APIs claras e suporte a múltiplos cenários de deployment. Requer Node.js 18+ e acesso ao MongoDB, com setup direto.

README

A tradução para o português deste README ainda está na fila.

LibreChat Analytics MCP Server

A dedicated MCP (Model Context Protocol) server that connects directly to your LibreChat MongoDB database to provide real-time analytics on token usage, model consumption, and user activity.

Tools

getModelUsage

Returns total token consumption broken down per model across your LibreChat instance.

No input required.

Example output:

{
  "gpt-4o": 120000,
  "claude-3-5-sonnet": 85000,
  "gemini-pro": 43000
}

getTokenUsage

Returns token usage aggregated over a specified time period.

Input:

ParameterTypeValues
periodstringdaily, monthly, quarterly, yearly

getTopUsers

Returns the top users ranked by total token consumption.

Input:

ParameterTypeDescription
limitnumberNumber of top users to return (e.g. 10)

getUsageSummary

Returns a full usage summary for a specified period including total tokens, total messages, and active users.

Input:

ParameterTypeValues
periodstringdaily, monthly, quarterly, yearly

Example output:

{
  "totalTokens": 248000,
  "totalMessages": 1340,
  "activeUsers": 38
}

Setup

Prerequisites

  • Node.js v18+
  • Access to your LibreChat MongoDB connection string

Installation

# Clone the repository
git "clone https://github.com/Avishkar-Kabadi-ODA/librechat-analytics-mcp"
cd librechat-analytics-mcp

# Install dependencies
npm install

Configuration

Create a .env file in the root directory:

MONGO_URI=mongodb+srv://<user>:<password>@<host>/?tls=true&retryWrites=false
DB_NAME=LibreChat
PORT=3100
VariableDescription
MONGO_URIYour LibreChat MongoDB connection string
DB_NAMEDatabase name — usually LibreChat
PORTPort to run the MCP server on

Start the server

npm start

Server will start at http://localhost:PORT/mcp


Connecting to LibreChat

Option 1 — LibreChat running locally via Docker

If LibreChat is running locally in Docker, connect directly using:

http://localhost:PORT/mcp

In LibreChat → Settings → MCP Servers → add:

URL: http://localhost:3100/mcp

Option 2 — LibreChat running on a remote server

You need to expose your local MCP server to the internet. Use either Cloudflare Tunnel or ngrok:

Using Cloudflare Tunnel (Recommended — free, no account required)

# Install cloudflared
winget install Cloudflare.cloudflared        # Windows
brew install cloudflare/cloudflare/cloudflared  # macOS

# Start tunnel
cloudflared tunnel --url http://localhost:3100

Cloudflare will generate a public URL like:

https://random-name.trycloudflare.com

Connect LibreChat using:

https://random-name.trycloudflare.com/mcp

Using ngrok

# Install ngrok from https://ngrok.com/download
# Then expose your port
ngrok http 3100

ngrok will generate a public URL like:

https://abc123.ngrok-free.app

Connect LibreChat using:

https://abc123.ngrok-free.app/mcp

⚠️ Free ngrok URLs change every time you restart. Use a paid plan or Cloudflare for a stable URL.


Option 3 — Deploy to cloud

Deploy this MCP server to any cloud platform for a permanent public URL:

Azure Container Apps

az acr build --registry <your-acr> --image librechat-analytics-mcp .
az containerapp create --name analytics-mcp --image <your-acr>.azurecr.io/librechat-analytics-mcp ...

Render (Free tier available)

  1. Push repo to GitHub
  2. Go to render.com → New Web Service
  3. Connect your repo
  4. Set environment variables in Render dashboard
  5. Deploy — Render provides a permanent public URL

Railway / Fly.io

Similar to Render — connect GitHub repo, set env vars, deploy.

Connect LibreChat using your deployed URL:

https://your-deployed-url.com/mcp

Adding to LibreChat

In your librechat.yaml:

mcpServers:
  analytics:
    url: https://your-mcp-url/mcp
    type: streamable-http

Or via LibreChat Admin Panel → MCP Servers → Add New.


Contributing

Contributions are welcome! If you'd like to add new tools, improve existing ones, or fix bugs:

  1. Fork the repository
  2. Create a new branch (git checkout -b feature/your-feature)
  3. Commit your changes (git commit -m 'Add your feature')
  4. Push to the branch (git push origin feature/your-feature)
  5. Open a Pull Request

For major changes, please open an issue first to discuss what you'd like to change.

License

MIT

Ver o original no GitHub

Bom para

Relacionados