Unichat MCP Server in Python

@MCP-Mirror

リポジトリ

基本情報

カテゴリ

その他

トランスポート

stdio

作者

MCP-Mirror

Unichat MCP Server in Python について

Mirror of

#unichat#mcp-server #python

Unichat MCP Server in Python の設定

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": {
    "amidabuddha_unichat-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "unichat-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

Unichat MCP Server in Python のツール

Send a request to unichat

概要

What is Unichat MCP Server in Python?

Unichat MCP Server in Python is an MCP server that sends requests to multiple AI providers (OpenAI, MistralAI, Anthropic, xAI, or Google AI) using a single tool or predefined prompts. It requires a vendor API key and is intended for developers integrating AI capabilities via the Model Context Protocol.

How to use Unichat MCP Server in Python?

Install by configuring Claude Desktop with the server command and environment variables, or by using npx -y @smithery/cli install unichat-mcp-server --client claude. Set the UNICHAT_MODEL (e.g., gpt-4o-mini) and UNICHAT_API_KEY (your vendor API key) in the environment. The server provides one tool (unichat) and four prompts (code_review, document_code, explain_code, code_rework).

Key features of Unichat MCP Server in Python

Single tool (unichat) to query multiple AI vendors.
Predefined prompts for common coding tasks.
Supports OpenAI, MistralAI, Anthropic, xAI, and Google AI.
Configuration via environment variables (UNICHAT_MODEL, UNICHAT_API_KEY).
Can be installed from PyPI via uvx or directly from a local directory.
Automatable installation through Smithery.

Use cases of Unichat MCP Server in Python

Request AI completions from multiple providers using a unified MCP interface.
Review code for best practices and potential issues via the code_review prompt.
Generate documentation and docstrings with the document_code prompt.
Request detailed explanations of code snippets using explain_code.
Apply requested changes to existing code with the code_rework prompt.

よくある質問

Which AI models are supported?

A list of supported models is available in the project’s models file. The environment variable `UNICHAT_MODEL` selects one of them.

What API key is required?

You must set `UNICHAT_API_KEY` to a valid API key from the vendor corresponding to the chosen model.

How is the server deployed?

It runs as an MCP server over stdio. For local development, use `uv run unichat-mcp-server`. For published packages, use `uvx unichat-mcp-server`.

Can it be used with clients other than Claude Desktop?

The README only provides configuration examples for Claude Desktop, but as a standard MCP server it may work with any MCP‑compatible client.

How can I debug the server?

Use the MCP Inspector by running `npx @modelcontextprotocol/inspector uv --directory <source> run unichat-mcp-server`.

「その他」の他のコンテンツ

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov

Reactive Resume

amruthpillai

A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!

39.5K

Volley.sh

Volley

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi