mcp-client-for-ollama and arcade-mcp
About mcp-client-for-ollama
jonigl/mcp-client-for-ollama
A text-based user interface (TUI) client for interacting with MCP servers using Ollama. Features include agent mode, multi-server, model switching, streaming responses, tool management, human-in-the-loop, thinking mode, model params config, MCP prompts, custom system prompt and saved preferences. Built for developers working with local LLMs.
# Technical Summary Implements stdio, SSE, and HTTP transport protocols for MCP server communication with automatic reconnection and hot-reload capabilities during development. Built as a Python TUI using modern libraries (Typer, Rich, Textual) that connects Ollama models—both local and cloud-hosted—to MCP tool ecosystems for agentic workflows with iterative tool execution loops. Supports cross-language servers (Python/JavaScript), integrates Claude's native MCP configurations via auto-discovery, and provides safety mechanisms like human-in-the-loop approval gates before tool execution.
About arcade-mcp
ArcadeAI/arcade-mcp
The best way to create, deploy, and share MCP Servers
Provides a Python framework with MCPApp decorator-based tool definition, supporting both stdio and HTTP transports with built-in secret/OAuth token injection at runtime—keeping credentials inaccessible to LLMs and clients. Includes a CLI scaffolding tool (`arcade new`) that generates complete project boilerplate with pyproject.toml configuration, and integrates with Claude Desktop, Cursor, and VSCode through transport-specific connectors.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work