Github

Name: Github
Author: dre1144

@dre1144

リポジトリ

Github について

概要はまだありません

#finance #portfolio-analysis #recommendation-agent

基本情報

カテゴリ

金融とコマース

トランスポート

stdio

公開者

dre1144

投稿者

Andrey

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "github": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "SUPABASE_URL",
        "-e",
        "SUPABASE_SERVICE_KEY",
        "-e",
        "ENCRYPTION_KEY",
        "-e",
        "MCP_API_KEY",
        "-e",
        "MCP_API_URL",
        "-e",
        "GITHUB_PERSONAL_ACCESS_TOKEN",
        "mcp/finance-portfolio-agent"
      ],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is MCP Finance Agent?

MCP Finance Agent is a Model Context Protocol agent for analyzing financial portfolios and generating recommendations. It integrates with the Tinkoff Invest API to provide portfolio analysis, risk assessment, and optimization suggestions.

How to use MCP Finance Agent?

Clone the repository, install dependencies with Poetry, configure environment variables in a .env file, then run the agent using poetry run python -m src.main. The API becomes available at http://localhost:8000.

Key features of MCP Finance Agent

Integration with Tinkoff Invest API
Portfolio and risk analysis
Recommendation generation for portfolio optimization
Real-time market data monitoring
Efficiency metric calculations

Use cases of MCP Finance Agent

Automated portfolio risk assessment
Personalized portfolio rebalancing recommendations
Continuous monitoring of financial positions

FAQ from MCP Finance Agent

What runtime does MCP Finance Agent require?

Python 3.11+ and Poetry for dependency management are required.

How do I configure MCP Finance Agent?

Create a .env file based on .env.example and set the required environment variables (e.g., API keys for Tinkoff Invest).

How is the agent tested?

Run pytest to execute all tests with coverage reporting. Specific tests can be run individually, and tests can be filtered by marker (e.g., -m "not integration").

What is the structure of the project?

Source code lives in src/ with subdirectories for the agent core, external services, and data models. Tests are in tests/ and documentation in docs/.