広告掲載
MCP.so
ログイン

LLM Bridge MCP Server

@sjquant

リポジトリ

基本情報

カテゴリ

AI とエージェント

ライセンス

MIT

トランスポート

stdio

作者

sjquant

LLM Bridge MCP Server について

A model-agnostic Message Control Protocol (MCP) server that enables seamless integration with various Large Language Models (LLMs) like GPT, DeepSeek, Claude, and more.

#mcp#openai#claude#llm#anthropic#model-context-protocol#mcp-server#pydantic-ai#deepssek-r1

LLM Bridge MCP Server 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": {
    "llm-bridge-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@sjquant/llm-bridge-mcp",
        "--client",
        "claude"
      ]
    }
  }
}

LLM Bridge MCP Server のツール

5

The text prompt to send to the LLM

Specific model to use (default: "openai:gpt-4o-mini")

Controls randomness (0.0 to 1.0)

Maximum number of tokens to generate

Optional system prompt to guide the model's behavior

概要

What is LLM Bridge MCP?

LLM Bridge MCP allows AI agents to interact with multiple large language models through a standardized interface. It leverages the Message Control Protocol (MCP) to provide seamless access to different LLM providers, making it easy to switch between models or use multiple models in the same application.

How to use LLM Bridge MCP?

Configure API keys in a .env file (or pass them as environment variables) and then add a server entry to your Claude Desktop or Cursor configuration. The server exposes a single tool, run_llm, which accepts a prompt, optional model name (default openai:gpt-4o-mini), temperature, max tokens, and system prompt.

Key features of LLM Bridge MCP

  • Unified interface to OpenAI, Anthropic, Google, and DeepSeek models.
  • Built with Pydantic AI for type safety and validation.
  • Supports customizable temperature and max tokens parameters.
  • Provides usage tracking and metrics.
  • Simple single-tool API: run_llm.

Use cases of LLM Bridge MCP

  • Let an AI agent query multiple LLM providers from one integration.
  • Switch models in a running application without reconfiguration.
  • Compare outputs from GPT, Claude, Gemini, and DeepSeek in the same workflow.
  • Build a multi-model assistant that can choose the best model per task.

よくある質問

1

Which LLM providers are supported?

OpenAI (GPT models), Anthropic (Claude models), Google (Gemini models), and DeepSeek.

2

How do I install LLM Bridge MCP?

You can install it automatically via Smithery (`npx -y @smithery/cli install @sjquant/llm-bridge-mcp --client claude`) or manually by cloning the repository and following the manual installation steps.

3

How do I configure API keys?

Create a `.env` file in the root directory with keys like `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GOOGLE_API_KEY`, and `DEEPSEEK_API_KEY`. Alternatively, pass them as environment variables in the MCP server configuration.

4

How do I use LLM Bridge MCP with Claude Desktop or Cursor?

Add a server entry to your Claude Desktop configuration or `.cursor/mcp.json` file, specifying the command `uvx` and the package name `llm-bridge-mcp`, along with the required API keys.

5

What does the "spawn uvx ENOENT" error mean?

It means the system cannot find the `uvx` executable in your PATH. Find the full path to `uvx` using `which uvx` (macOS/Linux) or `where.exe uvx` (Windows), then update your MCP server configuration to use that full path.

コメント

「AI とエージェント」の他のコンテンツ

Plainpaper

Plainpaper

The AI marketing canvas for Claude and MCP agents. Your agent drafts campaigns as cards on a shared board, briefed by your brand rules, approved by you.

bsv.cx

Andy Rowe

Timestamp and verify evidence on-chain. Let your agent prove what it saw and when. bsv.cx gives an AI agent verifiable memory. Timestamp a hash of any output, archive a web page as it looked, or fetch a URL with a recei

Rumoro

@Rumoro

Social listening for AI agents. Rumoro's remote MCP server lets Claude, Cursor, Codex and other MCP clients search and sort mentions of your product, your competitors and your topics on Reddit, X, Hacker News, GitHub, Yo

CrawlForge MCP Server

mysleekdesigns

28-tool MCP server for web scraping, crawling, deep research & autonomous extraction — clean Markdown & structured JSON for Claude, Cursor & any MCP client. 1,000 free one-time credits, local-Ollama LLM support.

2

Nero AI Image Processing

nero-com

We provide AI-powered tools for image enhancement, upscaling, background removal, photo restoration, colorization, denoising, and compression. All features are available through our API documentation.

広告掲載