Otter Bridge MCP Server

@typangaa

代码仓库

基本信息

分类

开发工具

传输方式

stdio

作者

typangaa

提交者

Alex Pang

关于 Otter Bridge MCP Server

OtterBridge is a lightweight, flexible server for connecting applications to various Large Language Model providers. Following the principles of simplicity and composability outlined in Anthropic's guide to building effective agents, OtterBridge provides a clean interface to LLMs

#otterbridge#llm-connector#server

Otter Bridge MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "otterbridge": {
      "command": "uv",
      "args": [
        "run",
        "server.py"
      ]
    }
  }
}

Otter Bridge MCP Server 工具

Send messages to LLMs and get AI-generated responses

Retrieve information about available language models

概览

What is Otter Bridge?

Otter Bridge is a lightweight, flexible MCP server that connects applications to various Large Language Model providers. It provides a clean, composable interface for interacting with LLMs, currently supporting Ollama with planned expansion to ChatGPT and Claude. Designed for developers building agent workflows, it follows Anthropic’s principles of simplicity and composability.

How to use Otter Bridge?

Install prerequisites (Ollama and uv), clone the repository, install dependencies with uv add -r requirements.txt, copy .env.example to .env, and configure environment variables. Start the server manually with uv run server.py or automatically via MCP clients like Claude Desktop (add the server to the desktop configuration). Two tools are available: chat (send messages to LLMs) and list_models (retrieve available model information).

Key features of Otter Bridge

Provider-agnostic design (currently Ollama, ChatGPT and Claude planned)
Simple, composable architecture inspired by effective agent patterns
Lightweight server built with FastMCP
Easy model management and retrieval of model capabilities
Configurable via environment variables (OLLAMA_BASE_URL, DEFAULT_MODEL)

Use cases of Otter Bridge

Integrate a local LLM (Ollama) into any MCP-compatible application
Build custom agent workflows that switch between providers as they become available
Prototype LLM-powered features using a fast, local server before moving to cloud APIs
Seamlessly connect Claude Desktop with Ollama models for on-premise AI assistance

常见问题

What LLM providers does Otter Bridge support?

Currently, Ollama is supported. Support for ChatGPT is planned for Q2 2025 and Claude for Q3 2025.

What are the prerequisites for running Otter Bridge?

You need Ollama installed and running on the default port, and the `uv` Python package manager installed.

How do I integrate Otter Bridge with Claude Desktop?

Add the following configuration to your Claude Desktop config file (adjust the path to match your local installation): `{"otterbridge": {"command": "uv", "args": ["--directory", "<path-to-otterbridge>", "run", "server.py"]}}`.

What environment variables can I configure?

- `OLLAMA_BASE_URL`: URL of the Ollama server (default: http://localhost:11434) - `DEFAULT_MODEL`: Default model to use (default: llama3.3)

What tools does Otter Bridge expose?

It exposes two tools: `chat` for sending

开发工具分类下的更多 MCP 服务器

ArtiBot Cloud MCP - Zhijiangyun 智匠云

bofob

ArtiBot Cloud MCP (Zhijiangyun 智匠云) — cloud MCP server for robot manipulation data & GPU workloads. 15 tools: robot fleet management, data collection & video recording, post-processing pipelines, and benchmark evaluation

Mcp-Warden

ermoluk

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.

offprompt

MCP server that lets Claude Code, Cursor or Codex ask you for API keys on a local page and write them to .env. The agent only ever sees key names.

FARPY

mangomunchr

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

Heard

Let your cloud agents talk to you. Grok Bot, Meta Muse, ChatGPT, Claude, Devin and others report progress to Heard, which reads it out loud on your Mac, and they pick up the replies you send back.