convez.ai
MCP ServerDados & AnalyticsAvançado
bitlabsdevteam

SnowFlask Managed MCP Server

bitlabsdevteam/SnowFlask_managed_MCP_SERVER

MCP server seguro do Snowflake com Cortex AI para agentes Claude

Ver no GitHub
InicialPluginScore · de 0 a 100

Servidor MCP que expõe dados governados do Snowflake e capacidades de IA Cortex para agentes Claude de forma segura e controlada. Implementa acesso somente leitura com autenticação bearer token e lista de permissões para datasets específicos.

Instalação

$npx convez-ai bitlabsdevteam/SnowFlask_managed_MCP_SERVER

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
28%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
1

Casos de uso

  • Conectar Claude a data warehouses Snowflake com controle de acesso fino
  • Permitir agentes IA fazer buscas semânticas e consultas SQL governadas
  • Integrar capacidades de IA Cortex (completions e retrieval) em sistemas Enterprise

Veredito

MCP server robusto e bem arquitetado para expor Snowflake a agentes Claude com segurança em primeiro lugar. Excelente para cenários enterprise onde controle de acesso e governança são críticos; requer conhecimento sólido de Snowflake e infraestrutura containerizada.

README

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

Snowflake Managed MCP Server

A lean, security-first MCP server that exposes governed Snowflake data and Cortex AI to AI agents (e.g. agents built in Google Gemini Enterprise). See CLAUDE.md for the full architecture and security model.

Read-only by design. No DDL/DML. Access is restricted to an allow-list of governed views and enforced again by a least-privilege Snowflake role.

Tools

ToolWhat it does
list_datasetsList allow-listed governed datasets agents may query
describe_datasetColumn names/types for one allow-listed dataset
run_queryRun a single read-only SELECT/WITH query (row-capped, timed out)
cortex_searchSemantic retrieval over an approved Cortex Search service
cortex_completeLLM completion via SNOWFLAKE.CORTEX.COMPLETE (allow-listed models)

Cortex Analyst (NL→SQL) is a planned addition; it uses a REST endpoint and was left out of v0.1 to avoid shipping an untested HTTP client. The SQL-based Cortex tools above cover retrieval and completion today.

Transports

  • stdio — stdin/stdout for a trusted local agent process (no network auth).
  • sse — HTTP + Server-Sent Events, protected by bearer-token authentication (MCP_SSE_BEARER_TOKENS). Refuses to start if no tokens are configured.

Setup

python -m venv .venv && source .venv/bin/activate
pip install -e .
cp .env.example .env   # then fill in Snowflake creds, allow-list, and limits

Run

# Local trusted agent (stdio)
python -m snowflake_mcp --transport stdio

# Network agents (authenticated SSE)
python -m snowflake_mcp --transport sse
# Agents connect to http://$MCP_SSE_HOST:$MCP_SSE_PORT/sse with header:
#   Authorization: Bearer <one of MCP_SSE_BEARER_TOKENS>

Run with Docker (primary deployment)

The server is operated as a container service over SSE. Secrets are injected at runtime from .env — never baked into the image.

cp .env.example .env   # fill in creds, MCP_SSE_BEARER_TOKENS, and MCP_BEARER_TOKEN
docker compose up --build server     # start the MCP server (SSE on 127.0.0.1:8080)
docker compose run --rm client       # run the example client against it

The image runs as a non-root user and publishes only to 127.0.0.1. In production, place the SSE port behind a gateway enforcing mTLS/OAuth.

Test

pip install pytest
pytest                     # all guardrail tests
pytest tests/test_security.py::test_rejects_dangerous_sql   # a single test

Security notes

  • Set MCP_ALLOWED_OBJECTS to fully-qualified governed views only — an empty allow-list means the server can read nothing (safe default).
  • Prefer Snowflake key-pair auth; password is a dev-only fallback.
  • The Snowflake role in SNOWFLAKE_ROLE must be least-privilege (read-only on governed views). MCP guardrails are defense-in-depth, not a substitute for it.
Ver o original no GitHub

Bom para

Relacionados