Microsoft Teams MCP is an MCP server that integrates Microsoft Teams with AI assistants through the Model Context Protocol. Built by InditexTech, it connects directly to Microsoft Entra ID and Teams channel APIs, allowing developers, team leads, and automated agents to interact with workplace conversations programmatically. By exposing core Teams capabilities as standardized tools, the server enables language models to read incoming channel messages, inspect thread replies, and retrieve channel membership rosters. It also permits assistants to publish new conversation threads with custom titles, contribute responses to active discussions, and mention specific colleagues directly within messages. Designed for enterprise collaboration and support workflows, Microsoft Teams MCP supports both local stdio execution and streamable-http deployment modes, making it adaptable to local desktop environments or containerized network services.
Category: Communication & Messaging
Tags: chat, collaboration, messaging, microsoft teams
bash git clone https://github.com/InditexTech/mcp-teams-server cd mcp-teams-server 2. Set up Python 3.12 and dependencies using uv: bash uv venv uv sync --frozen --all-extras --dev 3. Configure required environment variables in your environment or a .env file: TEAMS_APP_ID, TEAMS_APP_PASSWORD, TEAMS_APP_TENANT_ID, TEAM_ID, and TEAMS_CHANNEL_ID. 4. Run the server using uv run mcp-teams-server for default stdio transport, or run uv run mcp-teams-server --transport streamable-http for HTTP mode. 5. Alternatively, pull and run the pre-built Docker image: bash docker run --env-file .env -it ghcr.io/inditextech/mcp-teams-server:latest 6. Add the executable command and environment parameters to your MCP client configuration (such as Claude Desktop or Cursor) as detailed in the project repository.Part of MCP Servers
You can clone the repository from GitHub and set up the dependencies using the uv package manager with Python 3.12. Alternatively, you can use the pre-built Docker container image from GitHub Packages by running docker pull ghcr.io/inditextech/mcp-teams-server:latest and passing your environment variables.
Microsoft Teams MCP allows AI clients to interact with Microsoft Teams channels. Key capabilities include reading channel messages, fetching full thread replies, listing channel members, starting new discussion threads with subject titles, and replying to existing threads while mentioning specific users.
The server supports standard input and output (stdio) as the default transport mechanism for local clients. It also supports streamable-http for HTTP-based remote integrations and retains support for legacy SSE transports for backwards compatibility with older MCP clients.
You must configure Microsoft Entra ID and Teams credentials via environment variables. These include TEAMS_APP_ID, TEAMS_APP_PASSWORD, TEAMS_APP_TENANT_ID, along with the specific TEAM_ID and TEAMS_CHANNEL_ID you want the server to access.
Yes, Microsoft Teams MCP is open source software. It is released under the Apache-2.0 open-source license and maintained on GitHub by InditexTech.