Connects an AI assistant to a browser chat in the Google Gemini web app, allowing it to read and send messages in a user-selected conversation using the existing browser login.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@agent-messaging-mcp Look at the active chat window and summarize the latest messages."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Connect an AI assistant to a browser chat you explicitly select and approve, using your existing browser login.
You choose the conversation and approve each assistant once. That assistant can then read and send in that chat until its connection ends. Your login stays in your browser: no provider API keys or exported credentials are needed.
Reply waiting uses page completion indicators where available and quiet-period estimates otherwise.
Experimental prototype. Not ready for everyday use or personal conversations. Sending is tested with local and simulated chats; sending in the actual Google Gemini web app is not yet verified. Messages observed on the page are not proof of delivery.
DEVELOPMENT.md: setup, usage and technical reference.
DESIGN.md: architecture and design rationale.
HANDOFF.md: development checkpoints, evidence and handover.
AGENTS.md: project workflow for coding agents.
Related MCP server: AgentBridge
See LICENSE.
This server cannot be deployed
Agent communication platform for agent to agent messaging via MCP. Messages, channels, skills.
AI agent messaging with durable inboxes, contacts and tasks. Connect through MCP with OAuth.
Surface customer & prospect context from Slack, email, transcripts and tickets in any MCP client.
Agent-to-agent messaging: directory, public lobby, DMs, channels, search. Stateless MCP + REST.
