home / mcp / google chat mcp server
Google Chat MCP server that lets AI assistants like Claude and Cursor participate directly in team conversations - search messages, help teammates, share files, and coordinate across chat platforms.
Configuration
View docs{
"mcpServers": {
"siva010928-multi-chat-mcp-server": {
"command": "python",
"args": [
"-m",
"src.server",
"--provider",
"google_chat",
"-local-auth"
]
}
}
}You can run a single AI-powered MCP server that connects your AI assistant to multiple chat platforms, including Google Chat, while keeping conversations and data under your organizationโs control. This server provides a unified interface to interact with team chats across providers, enabling cross-platform collaboration and context-aware assistance from your AI.
You use the MCP server by starting a local, multi-provider agent that talks to Google Chat and optionally other chat platforms at the same time. The Google Chat provider is production-ready and can operate alongside other providers through a unified interface. To begin, authenticate with Google Chat, then run the MCP server so your AI assistant can send, search, summarize, and attach content across your configured chat spaces. You can issue prompts that target multiple spaces or platforms in one go, and the AI will perform actions in each provider as needed.
Prerequisites you need before installing:
- Python 3.9+
- UV Package Manager (recommended)
- Google Cloud Project with Google Chat API enabled
- MCP Client (Claude Desktop, Cursor, or other MCP-compatible AI assistant)
Follow these steps to install and set up the server locally.
Configuration and runtime notes: The Google Chat MCP support is designed for local or on-premises deployments to maximize security and data privacy. You prepare a Google Cloud project, enable the Google Chat API, and complete the configuration in the Google Cloud Console (app name, logo, etc.). You then download credentials.json to the Google Chat provider directory before starting the server with local authentication.
Running multiple providers: You can run the Google Chat provider together with additional providers such as Slack or Teams. Each provider runs in its own server instance, yet all share a unified interface so your AI assistant can operate across platforms in parallel. This enables workflows like incident response, knowledge consolidation, and cross-platform team coordination.
Troubleshooting tips: Ensure Google Chat API scopes are granted, credentials.json is in the correct location, and network connectivity allows access to Google services. If authentication fails, re-run the local authentication step to refresh credentials.
Security and usage policies: Keep sensitive conversations within your organization. Use on-premises LLMs or locally hosted agents to maintain control over data and access policies. Implement organization-specific security measures and monitor data flows to meet compliance requirements.
List Google Chat spaces the AI can access.
Add or remove members from a Google Chat space.
Retrieve participants in a chat conversation.
Provide a concise summary of a conversation.
Send a message to a chat space or thread.
Reply within a specific message thread.
Update an existing chat message.
Delete a chat message.
React to a message with an emoji.
Fetch details about a specific chat message.
Search messages across spaces.
Find messages where you were mentioned.
Retrieve your user profile information.
Fetch information about a user by ID.
Get a message with sender details.
List messages with sender details.
Upload an attachment to a chat.
Send a fileโs content as a message.
Share formatted file content in a message.
Send multiple messages in a single operation.