Autopilot Monitor MCP
Autopilot Monitor ships an MCP server for investigating Windows Autopilot and Intune failures across sessions, ESP and apps.
- 95 ★ on the repository
- Last commit
- AGPL-3.0
Extended access
Why this level
- Device and application data of the organisation is passed to the agent and the hosted server.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/autopilot-monitorDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Device and application data of the organisation is passed to the agent and the hosted server.
This is third-party code. Review the repository files before installing.
What it does
Autopilot Monitor records Windows Autopilot and Microsoft Intune enrollment events while they run: app installs, policies, device state and diagnostics. The hosted MCP server at `https://mcp.autopilotmonitor.com/mcp` exposes this data to an agent. An agent can answer questions such as which enrollments failed in the last day and why, which app install stretched the ESP, and which device models fail most. Sign-in uses a work account, and access is scoped to your tenant exactly as in the portal.
Who it is for. Intune and Windows Autopilot administrators who investigate deployment failures across a device fleet.
Good fit when
- A deployment is stuck and the cause needs an event timeline.
- The ESP phase dragged and the responsible app or policy must be found.
- Failures repeat on one device model.
- You must check whether the fleet is affected by a specific vulnerability.
Not a fit when
- The organisation does not use Windows Autopilot or Intune.
- Device data cannot be sent to a cloud service.
- You need an offline tool without a hosted server.
Example request
Show failed enrollments from the last day, group them by cause and find the app that stretched the ESP.MCP
- Transport
- http
- Authentication
- OAuth
Limitations
The MCP works only on top of the platform: it needs the device agent deployed and an activated Autopilot Monitor organisation. The server is hosted, so fleet telemetry reaches the project's cloud. The licence is mixed, AGPL-3.0 for server parts and MIT for the device agent. The repository contains the device agent, backend, web dashboard and the MCP server itself, but using the MCP needs an activated organisation in the cloud service. Self-hosting the whole platform is described in the project documentation and needs separate infrastructure.
How to disable. Remove the connector from the AI client and revoke access in your account settings. Remove the device agent and organisation data following the project documentation.
FAQ
Which client works?
Claude Desktop, VS Code and any MCP client that supports Streamable HTTP with OAuth.
Does the server store credentials?
Per the README the server stores no credentials; sign-in goes through your work account.
Is the platform paid?
The Community plan is free, and organisations complete a short activation after first sign-in.
Which licence applies?
Mixed: AGPL-3.0 for server parts and MIT for the device agent.
What can be asked?
For example, failed enrollments in the last day, the cause of one session's failure, long ESP phases and device models with high failure rates.
Discussion
No questions or reviews yet.
Related
MCP serversThe built-in OneUptime MCP: an agent works incidents, reads logs and traces, acknowledges alerts and sets up monitors.
An agent searches AWS CloudWatch logs, summarizes activity and correlates events across services.
Console Automation MCP gives an agent local shells and SSH sessions: input, output, error detection, long processes
Measure MCP: an agent reads mobile app crashes, ANRs, traces, sessions and bug reports and looks for the root cause.
Datadog's official managed MCP server: logs, metrics, traces and incidents right in your agent
Datadog MCP Server (winor30)
Datadog MCP Server
Independent Datadog MCP server: an agent reads incidents, monitors, logs, metrics, traces and RUM and schedules downtimes.
README in short
The page describes platform features, the kinds of problems it helps investigate, the MCP server address, documentation and the split licensing.