MCP Shared Library

Shared models and services for MCP (Model Context Protocol) components.

Overview

This library provides common data models, services, and utilities used by:

• mcp_local_repo_analyzer - Git repository analysis tools
• mcp_pr_recommender - PR recommendation engine

Installation

cd mcp_shared_lib
poetry install

Usage

Models

from mcp_shared_lib.models import FileStatus, WorkingDirectoryChanges, RiskAssessment

# Create a file status
file_status = FileStatus(
    path="src/main.py",
    status_code="M",
    lines_added=10,
    lines_deleted=5
)

Services

from mcp_shared_lib.services import GitClient, ChangeDetector
from mcp_shared_lib.config import GitAnalyzerSettings

# Setup services
settings = GitAnalyzerSettings()
git_client = GitClient(settings)
change_detector = ChangeDetector(git_client)

# Use services
repo = LocalRepository(path=".", name="my-repo", current_branch="main", head_commit="abc123")
changes = await change_detector.detect_working_directory_changes(repo)

Architecture

mcp_shared_lib/
├── models/          # Data models
│   ├── git/         # Git-related models
│   ├── analysis/    # Analysis models
│   ├── pr/          # PR models (Phase 2)
│   └── base/        # Base classes
├── services/        # Business logic services
│   ├── git/         # Git services
│   └── pr/          # PR services (Phase 2)
├── tools/           # Tool base classes
├── utils/           # Utility functions
└── config/          # Configuration management

Development

Running Tests

poetry run pytest

Code Formatting

poetry run black .
poetry run isort .

Type Checking

poetry run mypy .

Migration Status

• ✅ Phase 1: Foundation setup (models from mcp_local_repo_analyzer)
• ⏳ Phase 2: PR models migration (from mcp_pr_recommender)
• ⏳ Phase 3: Service migration
• ⏳ Phase 4: Repository updates
• ⏳ Phase 5: Testing & cleanup