About M365 Roadmap
M365 Roadmap is a Python MCP server for querying Microsoft 365's public roadmap. It helps AI agents and organizations filter upcoming and recent product changes, including availability across government cloud instances.
Key features
- 01Search roadmap by keyword
- 02Filter by product and status
- 03Filter by cloud instance and platform
- 04Look up feature details by ID
- 05Find recent or modified entries
- 06Filter by release phase and dates
Use cases
- 01Check Teams features currently rolling out
- 02Check Copilot availability in GCC High
- 03Review OneDrive features for GCC High
- 04Find roadmap additions from the last 30 days
MCP support
Exposes a single `search_roadmap` tool to MCP clients. It supports combined filters and retrieval of feature metadata from Microsoft's public roadmap API.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
1Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| limit | integer | No | Maximum number of results to return (default: 10, max: 100). Ignored when feature_id is provided. |
| query | any | No | Optional keyword to match against title and description (case-insensitive). |
| status | any | No | Optional status filter. Valid values: In development, Rolling out, Launched. |
| product | any | No | Optional product tag to filter by (case-insensitive partial match, e.g. "Teams" matches "Microsoft Teams"). |
| platform | any | No | Optional platform filter (case-insensitive partial match). |
| feature_id | any | No | Optional roadmap ID to retrieve a single specific feature. When provided, all other filters are ignored. |
| preview_date | any | No | Optional preview availability date filter (partial string match against previewAvailabilityDate, e.g. "2026-07"). |
| rollout_date | any | No | Optional rollout date filter (partial string match against generalAvailabilityDate, e.g. "2026-03"). |
| release_phase | any | No | Optional release phase filter (case-insensitive partial match). |
| cloud_instance | any | No | Optional cloud instance filter (case-insensitive partial match, e.g. "GCC" matches "GCC", "GCC High" matches "GCC High"). |
| include_facets | boolean | No | When True, includes taxonomy facets (products, statuses, release_rings, platforms, cloud_instances) with occurrence counts in the response. Use with limit=0 to get only facets without features. Facets are computed from matched results after filters are applied. |
| added_within_days | any | No | Optional number of days to look back for recently added features (clamped to 1-365). Only features with a created date within this window are returned. |
| modified_within_days | any | No | Optional number of days to look back for recently modified features (clamped to 1-365). |
Details
From the project READMEDisclaimer: This is an independent, self-built project and is not an official Microsoft tool or service.
M365 Roadmap MCP Server
A Python-based MCP (Model Context Protocol) server that enables AI agents to query the Microsoft 365 Roadmap programmatically.
Strategic Rationale
For organizations relying on Microsoft 365, Teams, or SharePoint, the "Roadmap" is the single source of truth for upcoming changes. However, navigating the roadmap website manually is cumbersome and disconnected from technical planning workflows. "When is Copilot coming to GCC High?" is a question that affects multi-million dollar contracts and deployment schedules.
Existing research indicates that while RSS feeds exist, there is no tool that allows an AI agent to structurally query this data to answer complex filtering questions. A "Roadmap Scout" MCP server empowers the Agent to act as a release manager, proactively identifying features that enable new capabilities or threaten existing customizations.
Prompt Examples
Once connected to an MCP client, you can ask questions like:
- Search by product and status: "What Microsoft Teams features are currently rolling out?"
- Check government cloud availability: "Is Copilot available for GCC High yet?"
- Find recent additions: "Show me everything added to the M365 roadmap in the last 30 days"
- Get feature details: "Tell me more about roadmap feature 534606"
- Government cloud planning: "My agency is on GCC High. Which OneDrive features can we expect?"
Installation
Prerequisites
- Python 3.11+
- An MCP-compatible client (Claude Desktop, Cursor, Claude Code, GitHub Copilot CLI, etc.)
From PyPI (recommended)
Using uvx (requires uv):
uvx m365-roadmap-mcp
To update to the latest version:
uvx m365-roadmap-mcp@latest
Install uv if you don't have it:
# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows (PowerShell)
irm https://astral.sh/uv/install.ps1 | iex
Or install with pip (no uv required):
pip install m365-roadmap-mcp
# Update to latest
pip install --upgrade m365-roadmap-mcp
Quick Setup
One-click install: Click VS Code badge for automatic setup (requires
uvinstalled) Manual install: See instructions below for Cursor, Claude Code, Copilot CLI, or Claude Desktop
Client Configuration
Running the server
uvx m365-roadmap-mcp
Or if installed with pip:
m365-roadmap-mcp
Claude Desktop
Add to your Claude Desktop MCP config:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Using uvx (recommended)
{
"mcpServers": {
"m365-roadmap": {
"command": "uvx",
"args": ["m365-roadmap-mcp"]
}
}
}
Using installed package
{
"mcpServers": {
"m365-roadmap": {
"command": "m365-roadmap-mcp"
}
}
}
Cursor
Option 1: One-Click Install (Recommended)
cursor://anysphere.cursor-deeplink/mcp/install?name=m365-roadmap-mcp&config=eyJjb21tYW5kIjogInV2eCIsICJhcmdzIjogWyJtMzY1LXJvYWRtYXAtbWNwIl19
Option 2: Manual Configuration
Add to your Cursor MCP config:
- macOS:
~/Library/Application Support/Cursor/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json - Windows:
%APPDATA%\Cursor\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json
Claude Code
claude mcp add --transport stdio m365-roadmap -- uvx m365-roadmap-mcp
GitHub Copilot CLI
Add to ~/.copilot/mcp-config.json:
{
"mcpServers": {
"m365-roadmap": {
"type": "stdio",
"command": "uvx",
"args": ["m365-roadmap-mcp"]
}
}
}
Features
Provides a single search_roadmap tool that handles all M365 roadmap queries. Combine any filters:
- Keyword search -- Find features by keyword in title/description
- Product filter -- Filter by product tag (Teams, SharePoint, etc.)
- Status filter -- Filter by status (In development, Rolling out, Launched)
- Cloud instance filter -- Filter by cloud instance (GCC, GCC High, DoD)
- Feature lookup -- Retrieve full metadata for a specific roadmap ID
- Recent additions -- List features added within the last N days
- Release phase filter -- Filter by release phase (General Availability, Preview, Targeted Release)
- Platform filter -- Filter by platform (Web, Desktop, iOS, Android, Mac)
- Rollout date filter -- Filter by general availability date (e.g., "2026-03", "2026")
- Preview date filter -- Filter by preview availability date (e.g., "2026-07", "2026")
- Recently modified -- List features modified within the last N days
Data Source
This MCP server pulls data from Microsoft's public roadmap API:
- API Endpoint:
https://www.microsoft.com/releasecommunications/api/v2/m365 - Authentication: None required (public endpoint)
- RSS Mirror:
https://www.microsoft.com/microsoft-365/RoadmapFeatureRSS(same data, RSS format)
This is the same data that powers the Microsoft 365 Roadmap website. The legacy endpoint (roadmap-api.azurewebsites.net) was retired in March 2025.
Coverage and Limitations
The API returns approximately 1,900 active features -- those currently In Development, Rolling Out, or recently Launched. This is a hard cap; older or retired features age out of the API and are no longer returned. The roadmap website may display historical features that are no longer present in the API.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.