Dive - AI Developer Tools Tool

Overview

Dive is an open-source, cross-platform MCP (Model Context Protocol) host desktop application that centralizes LLM-based assistant workflows, model management, and tool orchestration. It provides universal LLM support (OpenAI/ChatGPT-compatible models, Anthropic, Ollama and other OpenAI-compatible providers), built-in MCP tooling (fetch, filesystem, bash), and UI-driven management for adding, editing, and enabling/disabling MCP servers. Dive ships as both a modern Tauri build and a traditional Electron build so users can choose a smaller-footprint runtime or a more established Electron experience. ([github.com](https://github.com/OpenAgentPlatform/Dive)) Designed for developers and power users, Dive includes advanced API and model key management, custom system instructions, multi-language UI, auto-update capabilities, and an MCP Server Installer Agent that can help install and configure MCP servers automatically. It integrates with OAPHub (one-click managed MCP servers) to let teams use managed tool backends without local Python/Docker setup. Recent releases added file-path search, improved PATH handling on Linux/macOS, and UI navigation improvements. The project is distributed under an MIT license and maintains an active release cadence and community channels (Discord, Reddit). ([github.com](https://github.com/OpenAgentPlatform/Dive))

GitHub Statistics

• Stars: 1,734
• Forks: 163
• Contributors: 12
• License: MIT
• Primary Language: TypeScript
• Last Updated: 2026-02-13T05:33:15Z
• Latest Release: v0.13.2

The repository is actively maintained with a long commit history and frequent releases; the public GitHub page shows ~1.7k stars and 163 forks, an MIT license, and dozens of releases and contributors. The project exposes ~21 open issues and a small number of active pull requests, indicating steady community usage and triage activity. The maintainers publish both Tauri and Electron builds and maintain a separate MCP-host component, with recent commits as of February 2026. Overall the repo demonstrates healthy interest and a small but active contributor base. ([github.com](https://github.com/OpenAgentPlatform/Dive))

Installation

Install via npm:

git clone https://github.com/OpenAgentPlatform/Dive.git

cd Dive

npm install

npm run dev   # start Electron development build

cargo tauri dev   # start Tauri development build (requires Rust/cargo)

chmod +x Dive-*.AppImage && ./Dive-*.AppImage   # run Linux AppImage builds

paru -S dive-ai   # Arch Linux (AUR) example

Key Features

• Universal LLM support: OpenAI/ChatGPT-compatible, Anthropic, Ollama, and custom model IDs.
• Model Context Protocol host: run MCP servers in stdio or SSE mode for tool orchestration.
• Built-in tools: Fetch (web requests), File Manager (read/write), and Bash command execution.
• Dual architecture: Tauri (small installer) and Electron (stable AppImage/.dmg) distributions.
• Advanced API & model management: multiple API keys, model settings, custom system prompts.

Community

Dive has an engaged user and developer community (GitHub stars ~1.7k, forks ~163, active releases), official chat/discussion channels (Discord, Reddit), and documented development setup. The project has responded to multiple high-severity security reports—several CVEs related to deeplink/XSS handling were disclosed and fixed in subsequent releases—so users should run recent versions and follow security advisories. Overall, the community is small but active, with steady contributions and issue triage. ([github.com](https://github.com/OpenAgentPlatform/Dive))

Key Information

• Category: Developer Tools
• Type: AI Developer Tools Tool