Repolode
Submit
QuixiAI

QuixiAI/dolphin-mcp

A Python library and CLI that lets any supported LLM (OpenAI, Anthropic, Ollama, LMStudio) call tools on one or more MCP servers via natural language.

View on GitHub
Stars
533
Forks
59
Open issues
18
Last commit
23 days ago

What is it?

Dolphin MCP is a Python library and command-line tool for querying Model Context Protocol (MCP) servers through natural language. It connects to any number of configured MCP servers, discovers the tools they expose, and passes those tools to a language model so the model can call them while answering a question.

Supported model providers are OpenAI, Anthropic, Ollama, and LMStudio. Servers are declared in a JSON file with a command, arguments, and optional environment variables, and API keys are read from a .env file. The CLI offers a single-shot mode, an interactive chat mode that keeps context across turns, and a JSONL log of LLM interactions.

The package can be installed from PyPI, which provides the dolphin-mcp-cli command. The repository includes a demo SQLite server backed by a sample dolphin species database, created by a setup script.

Key features

  • 01Connects to multiple MCP servers at once and makes all their tools available to the model.
  • 02Supports OpenAI, Anthropic, Ollama, and LMStudio models.
  • 03Provides both a Python library API and the dolphin-mcp-cli command.
  • 04Offers single-shot and interactive chat modes.
  • 05Logs LLM interactions to a JSONL file for inspection.

Use cases

  • Querying a database or other data source through an MCP server using natural language.
  • Testing MCP servers with different LLM providers from the command line.
  • Embedding MCP tool use into a Python application via the library API.
  • Running local models with Ollama or LMStudio alongside MCP tools.

Quick start

  1. 1

    Install the package from PyPI

    pip install dolphin-mcp
  2. 2

    Set your OpenAI API key in a .env file (add OPENAI_API_KEY to the file)

  3. 3

    Run the CLI with a query

    dolphin-mcp-cli "Your query here"

Things to know

  • Requires an OpenAI API key or another supported provider key, plus configuring your own MCP servers.
  • The README's license section is an unfilled placeholder and the metadata lists no license.

Topics

#mcp#llm#function-calling#python#cli#openai#anthropic#ollama

FAQ

What is dolphin-mcp?
A Python library and CLI that lets any supported LLM (OpenAI, Anthropic, Ollama, LMStudio) call tools on one or more MCP servers via natural language. Dolphin MCP is a Python library and command-line tool for querying Model Context Protocol (MCP) servers through natural language. It connects to any number of configured MCP servers, discovers the tools they expose, and passes those tools to a language model so the model can call them while answering a question.
Is dolphin-mcp free to use? What license does it have?
dolphin-mcp has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
Is dolphin-mcp still maintained?
Actively maintained. The last commit to dolphin-mcp was on Sep 29, 2026.
What can I use dolphin-mcp for?
Querying a database or other data source through an MCP server using natural language. Testing MCP servers with different LLM providers from the command line. Embedding MCP tool use into a Python application via the library API. Running local models with Ollama or LMStudio alongside MCP tools.
How do I install dolphin-mcp?
Install the package from PyPI: pip install dolphin-mcp Set your OpenAI API key in a .env file (add OPENAI_API_KEY to the file) Run the CLI with a query: dolphin-mcp-cli "Your query here"
Where can I get dolphin-mcp?
The source code and installation instructions are on GitHub: https://github.com/QuixiAI/dolphin-mcp

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects