Skip to content

Dive AI

Dive AI is a streamlined chat client that treats MCP tool results as first-class visual elements. When the model calls an MCP tool, the result appears as an expandable card — not just raw text. This makes complex tool outputs (JSON, tables, structured data) readable and navigable within the conversation flow.


Dv
Dive AI
Open-Source · GitHub
TRANSPORT
Streamable HTTP ✓
PLATFORM
Windows · macOS · Linux
MCP VIA
Settings

Clean Tool Visualization in Chat

Dive AI keeps conversations readable even when heavy tool usage is involved. Each MCP tool call renders as a collapsible card showing the tool name, parameters, and response. Expand for details, collapse to keep the conversation flow clean. Complex JSON responses are formatted with syntax highlighting.

Features:

  • Tool result cards — structured display of MCP tool outputs
  • Multi-provider — supports major LLM providers
  • Syntax highlighting — code and JSON rendered with proper formatting
  • Conversation export — save conversations with tool call history
  • Desktop app — native cross-platform application
  • Minimal design — focused on conversation without UI clutter

Setup

1. Create a Token

In Vinkius Cloud, go to your server → Connection TokensCreate. Copy the URL.

2. Add MCP Server

Go to SettingsMCPAdd. Paste your Vinkius URL.

3. Chat with Visual Tool Cards

Ask questions. MCP tool results appear as formatted, expandable cards inline in the conversation.


FAQ

How does Dive AI display MCP tool results? As expandable cards in the conversation thread. Each card shows tool name, parameters sent, and formatted response data.

Does Dive AI support multiple MCP servers? Yes. Configure multiple servers. Tool cards are labeled by server for clarity.

Can I export conversations with tool calls? Yes. Exports include the full tool call history alongside messages.

Is Dive AI free? Open-source and free to use.