OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

Dive

Dive is an open-source desktop MCP host that connects function-calling language models with Model Context Protocol tools over stdio or SSE.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dive — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Dive when you want a desktop interface for connecting supported LLMs to MCP tools, choosing available tools, and working with built-in local utilities or configured MCP servers.

Who it is for

  • People who want to connect function-calling LLMs to MCP tools from a desktop interface.
  • Developers or technical users configuring local or managed MCP servers alongside model settings.

Documented uses

  • Connect an LLM that supports function calling to MCP tools.
  • Choose which individual MCP tools are enabled.
  • Use built-in Fetch, File Manager and Bash tools in model-assisted workflows.
  • Manage multiple API keys and switch models.
Before you choose

Requirements & limits

The repository lists Windows, macOS and Linux builds, with build choices differing by platform. Dive is open source under the MIT License. The documented OAPHub managed-server setup requires signing up with OAPHub.

  • The language model must support function calling.
  • The README lists compatibility with ChatGPT, Anthropic, Ollama and OpenAI-compatible models.
  • Python and Node.js setup requirements differ across operating systems and builds.
  • MCP server authentication is described as unstable and may require re-authorization.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Dive on Windows

Install the Windows version of the Dive desktop application.

Before you start

  • A Windows computer.
  • Choose either the Tauri or Electron version from the latest release.

Steps

  1. Open the latest Dive release page.

  2. Choose the Tauri installer or the Electron version.

  3. Launch the installer and complete the installation prompts.

  4. Launch Dive; Python and Node.js environments are downloaded automatically after launch.

Expected outcome

Dive is installed and its Python and Node.js environments are downloaded after the application is launched.

Guide limitations

  • The README describes Tauri as the recommended option with an installer under 30 MB; Electron is described as fully stable.
  • This procedure covers installation only, not connecting an LLM or configuring MCP servers.
Documentation 1 ↗
Source-based practical guide

Connect Dive to managed MCP servers through OAPHub

Connect Dive to OAPHub-managed MCP servers using a deep link or configuration file.

Before you start

  • Dive installed.
  • An OAPHub.ai account.

Steps

  1. Sign up at OAPHub.ai.

  2. Connect to Dive using a one-click deep link or configuration file.

  3. Use the managed MCP servers through Dive.

Expected outcome

Dive is connected to managed MCP servers through OAPHub, without requiring local Python, Docker, or other complex dependencies for this setup option.

Guide limitations

  • This procedure applies to OAPHub-managed servers; it does not explain configuration of other MCP servers.
  • The repository directs readers to MCP Servers Setup for more detailed instructions.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
10 October 2026
Original website title
OpenAgentPlatform/Dive · Official homepage title · retained in its source language
How we review sourcesSuggest a correction