MCPSERV.CLUB
joinly-ai

joinly.ai MCP Server

MCP Server

AI agents that join and interact in real‑time video calls

Active(80)
370stars
2views
Updated 13 days ago

About

joinly.ai’s MCP server lets AI agents enter meetings on Google Meet, Zoom, and Teams to listen, speak, and perform tasks via voice or chat, powered by modular LLM, STT, and TTS services.

Capabilities

Resources
Access data sources
Tools
Execute functions
Prompts
Pre-built templates
Sampling
AI model interactions

Joinly Demo

Overview

Joinly.ai is a Model Context Protocol (MCP) server that transforms ordinary video calls into collaborative, AI‑powered workspaces. By acting as a middleware layer between an AI assistant and the meeting platform, it grants the agent real‑time access to audio, chat, and screen content. This capability solves a common pain point for developers: enabling assistants to listen, understand, and act within the natural flow of a conversation without manual intervention or complex integration work.

At its core, Joinly exposes a set of meeting tools and resources through the MCP interface. These include speech‑to‑text (STT) engines, text‑to‑speech (TTS) backends, and context extraction utilities that convert the meeting’s audio stream into structured data. The server also offers a conversational flow controller that manages interruptions, speaker turns, and multi‑speaker dynamics, ensuring the AI’s responses feel natural rather than robotic. Developers can plug in any LLM provider—OpenAI, Anthropic, Ollama, or others—so the assistant’s intelligence can be tailored to specific use cases.

Key features that make Joinly valuable for AI workflows are:

  • Live Interaction: The agent can speak or type in real time, responding instantly to questions or commands.
  • Cross‑Platform Compatibility: Works seamlessly with Google Meet, Zoom, Microsoft Teams, and any browser‑based meeting service.
  • Modular Speech Services: Choose from Whisper/Deepgram for STT and Kokoro, ElevenLabs, or Deepgram for TTS.
  • Privacy‑First & Self‑Hosted: All data stays on the user’s infrastructure unless explicitly shared, addressing compliance and security concerns.
  • Bring‑Your‑Own LLM: The system is agnostic to the underlying language model, enabling experimentation and cost optimization.

Typical use cases span from meeting summarization (the agent captures key decisions and action items) to real‑time collaboration tools such as editing a Notion page or creating GitHub issues on the fly. In an enterprise setting, Joinly can act as a virtual meeting assistant that records compliance‑critical conversations, logs decisions, and triggers downstream workflows automatically.

By integrating Joinly into an AI pipeline, developers can elevate a standard video call into an intelligent workspace where assistants not only consume information but actively participate, augmenting productivity and reducing manual overhead.