** 서류** | PyPI 패키지

**MCP (모델 컨텍스트 프로토콜) 스쿼시 레이어 - LLM에 대한 레지스트리 드라이브 **

sktime의 기본 레지스트리와 semantics를 대형 언어 모델에 노출시키는 semantic 엔진은 다음을 가능하게합니다.

• ܧ Discover 유효한 자극자
• estReason estimator 기능에 대해
• 12.0 **컴포지트 ** 호환되는 estimators
• namExecute 실제 데이터 워크플로우

디자인 철학

이 MCP는 ** 문서 또는 정적 코드 분석이 아닙니다. programmatic 모형 사용법을 위한 **semantic 엔진 **입니다.

키 원칙

1. ** Truth의 소스로 sktime ** - AST 패싱 없음, 재포 지수 없음, 허리스틱 없음. 모든 구조는 all_estimators, estimator 태그 및 sktime’s API 계약에서 제공됩니다.

2. Registry-First - File → Class → Infer Relationships 대신, 우리는 Registry → Semantics → Safe Execution를 합니다.

3. ** 소수 MCP 표면 ** - LLM 요구 사항 만 노출 : Discovery, Description, Instantiation, Execution 및 모델 지속.

672️ 설치

Zero-install via uvx (권장)

uv가 설치되면 별도의 설치 단계가 필요하지 않습니다.

sktime-mcp

Read the Documentation | PyPI Package

MCP (Model Context Protocol) layer for sktime - Registry-Driven for LLMs

A semantic engine that exposes sktime’s native registry and semantics to Large Language Models, enabling them to:

• 🔍 Discover valid estimators
• 🧠 Reason about estimator capabilities
• 🔗 Compose compatible estimators
• ⚡ Execute real sktime workflows on real data

🎯 Design Philosophy

This MCP is not just documentation or static code analysis. It is a semantic engine for programmatic model usage.

Key Principles

1. sktime as Source of Truth - No AST parsing, no repo indexing, no heuristics. All structure comes from all_estimators, estimator tags, and sktime’s API contracts.

2. Registry-First - Instead of File → Class → Infer Relationships, we do Registry → Semantics → Safe Execution.

3. Minimal MCP Surface - Exposes only what an LLM needs: Discovery, Description, Instantiation, Execution, and model persistence.

🛠️ Installation

Zero-install via uvx (recommended)

If you have uv installed, no separate installation step is needed. Just update your MCP client config (see Connecting from an LLM Client below) and uvx will handle the rest automatically.

# Verify uv is available
uvx sktime-mcp --help

pip

pip install sktime-mcp

# With optional extras (SQL, forecasting models, file formats)
pip install "sktime-mcp[all]"

Development installation

git clone https://github.com/sktime/sktime-mcp
cd sktime-mcp
python3 -m pip install -e ".[dev]"

🐳 Docker

Run without installing anything locally (only Docker required):

# Build the image
docker build -t sktime-mcp .

# Run the MCP server (stdio transport)
docker run -i sktime-mcp

Or use Docker Compose:

docker compose build
docker compose run sktime-mcp

Claude Desktop — use Docker as the MCP server command:

{
  "mcpServers": {
    "sktime": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "sktime-mcp"]
    }
  }
}

Environment variables can be passed at runtime:

docker run -i -e SKTIME_MCP_LOG_LEVEL=DEBUG sktime-mcp

For a more detailed first-time setup flow, including MCP server verification and troubleshooting, see Beginner Setup.

🧭 Beginner Setup (First‑Time Users)

If you are new to sktime‑mcp or to MCP‑based workflows, this section provides a minimal starting point to help you verify that your setup is working correctly.

What is MCP?