Skip to main content

MCP server que recomenda modelos preditivos da árvore model-tree.

Project description

model-tree-mcp

Servidor MCP que recomenda modelos preditivos (ML, Deep Learning, Estatística Clássica) a partir da descrição de uma situação em linguagem natural.

A tool consulta um endpoint hospedado que faz a busca vetorial na árvore de 400+ modelos curados.

Uso (Claude Code / Claude Desktop)

Adicione ao seu config de MCP:

{
  "mcpServers": {
    "model-tree": {
      "command": "uvx",
      "args": ["model-tree-mcp"]
    }
  }
}

O uvx baixa e roda o pacote num ambiente isolado, sem instalação manual.

Apontar para outro endpoint

Por padrão a tool chama o endpoint público oficial. Para usar outro (ex.: um deploy próprio), defina a env var MODEL_TREE_API:

"env": { "MODEL_TREE_API": "https://seu-deploy.vercel.app/api/search" }

Tool e prompt

  • Tool search_models(situation: str, top_k: int = 8) — devolve os modelos mais próximos da situação descrita, cada um com seus campos (diff_siblings, strengths, weaknesses, recommended_for, not_recommended_for, keywords) e o stat_fit (perfil de encaixe estatístico: tipo/distribuição do target, regime n/p, tipos de feature, suposições, loss suportada, contraindicações).
  • Prompt analyze_dataset(data_path) — orquestra a recomendação a partir de um dataset local: o agente investiga (target, loss), faz a EDA (profunda no dado cru ou rasa numa EDA prévia, com os tokens do usuário) e recomenda 3-4 modelos com tradeoffs. Os dados crus nunca saem da máquina.

Desenvolvimento

uv run model-tree-mcp        # roda o server localmente (stdio)
uv build                     # gera o pacote distribuível

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

model_tree_mcp-0.1.0.tar.gz (4.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

model_tree_mcp-0.1.0-py3-none-any.whl (5.1 kB view details)

Uploaded Python 3

File details

Details for the file model_tree_mcp-0.1.0.tar.gz.

File metadata

  • Download URL: model_tree_mcp-0.1.0.tar.gz
  • Upload date:
  • Size: 4.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.23 {"installer":{"name":"uv","version":"0.11.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for model_tree_mcp-0.1.0.tar.gz
Algorithm Hash digest
SHA256 9a5841a27716bc5e2f6f6e5da8524cc2c809acbb7116143ab96118bc43493d03
MD5 50b61c5b634ba4b97f10f90d8fb13c16
BLAKE2b-256 cb80e380674e18cae29f438ad0b318a1b78bce2f60db1609cc620c8ef8d9e7ce

See more details on using hashes here.

File details

Details for the file model_tree_mcp-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: model_tree_mcp-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 5.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.23 {"installer":{"name":"uv","version":"0.11.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for model_tree_mcp-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 2d12cfa9abe517a99d10698435047af35d1c1f75a457c7f900cc63ab8e11f626
MD5 eeb0bd6fdaf0e5c7c047a0fd158c6dbd
BLAKE2b-256 165870abc716e974398dfc2665f8b117420b8dd80873ad336bd569bf1afb824f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page