mcp-registry/mcp-teams-server

    ==================
      
       /// MCP ///
      /// MCP ///
        
    ==================
        
    [server:online]
    [protocol:ready]

mcp-teams-server

by InditexTech

Python implementation of an MCP (Model Context Protocol) server that connects to Microsoft Teams to read, create and reply to channel messages, mentions users and lists members.

275
18
Specialized
GitHub

01

read messages

Capability to fetch and display messages from Microsoft Teams channels or threads

02

create messages

Capability to post a new message into a Microsoft Teams channel or thread

03

reply to messages

Capability to add a reply to an existing Microsoft Teams message or thread

04

mention members

Capability to mention specific Microsoft Teams users within a message

05

Start thread in channel with title and contents, mentioning users

Starts a new conversation thread in a channel, setting a title, body content and including user mentions

06

Update existing threads with message replies, mentioning users

Adds replies to an existing thread while optionally mentioning users

07

Read thread replies

Retrieves the list of replies in a specific conversation thread

08

List channel team members

Lists the members (users) that belong to a given Microsoft Teams channel

09

Read channel messages

Reads all messages present in a specified Microsoft Teams channel

Installation

1. Clone the repository
git clone https://github.com/InditexTech/mcp-teams-server.git cd mcp-teams-server
2. Create and activate a virtual environment (recommended)
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate
3. Install server dependencies
pip install -r requirements.txt # If the repo uses Poetry instead: # poetry install
4. Configure Microsoft-Teams / Azure credentials
# Set the following environment variables (obtained from Azure AD App Registration) export AZURE_CLIENT_ID="<app-id>" export AZURE_TENANT_ID="<tenant-id>" export AZURE_CLIENT_SECRET="<secret>" export TEAMS_BOT_ID="<bot-id>" # Optional – if bot identity is separate
5. Optional: edit config.yaml (or .env) to tweak port, logging and allowed scopes.
6. Launch the MCP server (default FastAPI + Uvicorn entry-point)
uvicorn mcp_teams_server.main:app --host 0.0.0.0 --port 8080
7. Verify the health-check endpoint
curl http://localhost:8080/health
8. Create a public HTTPS tunnel (for local testing)
ngrok http 8080 # Copy the public URL into your Teams bot configuration.
9. Deploy
# Container image (example) docker build -t inditex/mcp-teams-server . docker run -p 8080:8080 -e AZURE_CLIENT_ID=... inditex/mcp-teams-server

Documentation

License: Apache License 2.0
Updated 7/15/2025
mcp-teams-server - MCP Server Registry - Augment Code