mcpField
D
DialogBrain

DialogBrain

Remote
dialogbrain.comCommunication1Updated 24w ago

About DialogBrain

DialogBrain is a remote MCP server for teams using WhatsApp, Telegram, Instagram, email, live chat, and voice calls. It combines cross-channel inbox actions with workspace knowledge lookup and AI-agent orchestration.

Key features

  1. 01Read and search conversations across channels
  2. 02Send, forward, and delete messages
  3. 03Find and synchronize contacts
  4. 04Search workspace knowledge and graph relationships
  5. 05Trigger and review AI-agent actions
  6. 06Access file, calendar, call, and voice tools

Use cases

  1. 01Find unread messages across communication channels
  2. 02Reply or forward a message from an inbox
  3. 03Look up a policy or knowledge-graph entity
  4. 04Ask an AI agent to handle an inbound conversation

MCP support

Exposes 156 read/write tools through a remote MCP endpoint using Streamable HTTP and OAuth 2.0. MCP clients can access messaging, contacts, workspace knowledge, and other listed capabilities.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://api.dialogbrain.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

165
ParameterTypeRequiredDescription
channelenumNoFilter by channel. OMIT to search across all channels.telegram | email | whatsapp | sms | slack
limitintegerNoMaximum number of results to return
querystringYesName or username to search for (supports partial matches)

Details

From the project README

DialogBrain MCP

Remote MCP server that connects your AI agent to a unified inbox spanning WhatsApp, Telegram, Instagram, Email, Livechat, and Voice calls — plus a workspace knowledge graph and AI agent orchestration.

What you can do

  • Read messages across channels: messages.read_history, threads.list, threads.search
  • Send messages: messages.send, messages.forward, messages.delete
  • Manage contacts: contacts.find, contacts.discover, contacts.sync, contacts.profile
  • Query workspace knowledge: workspace.search, kg.find_entity, kg.get_relationships
  • Trigger AI agents: agents.ask, agents.simulate_inbound, agents.approve_draft
  • Files, calendar, calls, voice, browser automation, YouTube — see full tool list in Developer Docs.

Quick start

Claude Desktop / Claude.ai (web)

Add as a Custom Connector with URL https://api.dialogbrain.com/mcp. Sign in via OAuth when prompted.

ChatGPT (Apps)

In the ChatGPT Apps directory, search for "DialogBrain" (pending review). Or add as a Custom MCP Server with the same URL.

Cursor / generic MCP clients
{
  "mcpServers": {
    "dialogbrain": {
      "url": "https://api.dialogbrain.com/mcp",
      "transport": "streamable-http",
      "auth": "oauth2"
    }
  }
}

Authentication

OAuth 2.0 with Dynamic Client Registration (RFC 7591). Your MCP client registers automatically and is granted an access token scoped to one DialogBrain workspace. Revoke at any time from Settings → Developer in app.dialogbrain.com.

Tool annotations

Every tool exposes accurate MCP hint annotations:

  • readOnlyHint — true for read tools (threads.list, messages.read_history, kg.find_entity, …)
  • destructiveHint — true for delete tools (messages.delete, widgets.delete, …)
  • openWorldHint — true for tools that touch third-party APIs (messages.send, web.search, contacts.sync, …)
  • idempotentHint — derived from read-only/delete-by-id semantics

See server.json for the registry manifest and gemini-extension.json for Gemini CLI integration.

Privacy & support

License

MIT — see LICENSE.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.