# Microsoft Defender MCP connector

The Microsoft Defender connector brings alerts, devices, assessments and configuration scores into Claude, ChatGPT, Cursor, Microsoft Copilot and the Elaichi Agent, so each person can ask about threats and compliance inside their own Defender access.

Source: https://elaichi.ai/connectors/msdefender/

## Facts

| | |
| --- | --- |
| Application | Microsoft Defender |
| Category | Security |
| AI tools | 110 |
| Authentication | Connects over OAuth |
| Bring your own OAuth app | Yes |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, Microsoft Copilot, any MCP client, and the Elaichi Agent |
| Tools advertised by name | No. Connected tools are never listed one by one, however few there are. The endpoint advertises `search_tools` and `execute_tool` instead |

## What you can ask once Microsoft Defender is connected

- Show me open high severity alerts and the machines they hit
- Which devices have antivirus that is out of date?
- Summarize baseline compliance assessments failing this week

## Connect Microsoft Defender in Elaichi

This happens once for the organization, before any client is involved.

1. Open Connections, choose Add connection, and pick Microsoft Defender.
2. Optionally set Share with, then press Connect.
3. Supply your own Microsoft Defender app, then approve. Microsoft Defender needs an OAuth app you register yourself, so bring its client ID and secret. One person does this once for the organization.

Credentials are vaulted and nobody, including the AI, reads them back. The connection becomes a toolbox immediately, so you can curate which Microsoft Defender tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## Microsoft Defender MCP connector for Claude

Endpoint: https://api.elaichi.ai/mcp

1. Open Customize, then Connectors.
2. Press Add.
3. Name it, paste the MCP server URL, then Continue.
4. Sign in and approve.

On Team and Enterprise, an Owner adds it once. Everyone else turns it on for themselves.

## Microsoft Defender MCP connector for Claude Code

Endpoint: https://api.elaichi.ai/mcp

1. In a terminal, run `claude mcp add` with the endpoint.
2. Start Claude Code and run `/mcp`.
3. Pick elaichi, then sign in and approve in the browser.

```bash
claude mcp add --transport http elaichi --scope user https://api.elaichi.ai/mcp
```

Use `--scope project` instead to share it with the repository through `.mcp.json`.

## Microsoft Defender MCP connector for ChatGPT

Endpoint: https://api.elaichi.ai/mcp

1. Open Plugins, then press the + button.
2. Name it and paste the endpoint into Server URL.
3. Leave Authentication on OAuth, then tick the risk acknowledgement.
4. Press Create, then sign in and approve.

Works on the web. The plugin directory lives at chatgpt.com/plugins.

## Microsoft Defender MCP connector for Cursor

Endpoint: https://api.elaichi.ai/mcp

1. Open `~/.cursor/mcp.json`.
2. Add the endpoint under `mcpServers`.
3. Reload Cursor, then sign in and approve.

Set up per machine, so repeat it on each computer you work from.

## Microsoft Defender MCP connector for Microsoft Copilot

Endpoint: https://api.elaichi.ai/mcp

1. In Copilot Studio, open your agent, then Tools, Add a tool, New tool and Model Context Protocol.
2. Paste the endpoint into Server URL, choose OAuth 2.0, then Dynamic discovery, and press Create.
3. Create a new connection, sign in and approve, then press Add to agent.
4. Under Channels, open Teams and Microsoft Copilot and make the agent available in Microsoft Copilot.

Microsoft Copilot was called Microsoft 365 Copilot. Keep the tool on end-user credentials, so each person signs in as themselves.

## Connect Microsoft Defender to any MCP client

Endpoint: https://api.elaichi.ai/mcp

1. Add the endpoint as a remote MCP server.
2. Sign in and approve.

The Elaichi Agent already has these tools, with nothing to set up.

## What the consent screen decides

Every scope the client asks for starts ticked except "Delete things", which you tick yourself. Calling a Microsoft Defender tool needs "Use your connected apps". Over MCP there is no trusted place to confirm a write in the moment, so the consent screen is the standing approval rather than a formality. Think hard before granting Use your connected apps, which can also delete in connected apps and cannot be undone.

## What teams do with Microsoft Defender through Elaichi

### Triage this morning's alerts before standup

Security operations. Ask for the open Microsoft Defender alerts, which machines and users they touch, and which domains, files and IPs were involved. Update status and assignment on each alert, or on a whole batch, as you work through the queue.

### Check one suspicious domain across the fleet

Security operations. See every alert tied to a domain, which machines contacted it and how often, without opening a second console.

### Find devices with unhealthy antivirus

IT. Pull device antivirus health from Microsoft Defender and ask which machines are out of date or not reporting, then hand the list to the desk.

### Pull baseline compliance before the audit

Compliance. List baseline compliance assessments against your baseline profiles and configurations, then export them for the auditor in one ask.

### Review browser extensions and their permissions

IT. See which browser extensions are installed across devices and what permissions they ask for, and export the assessment for review.

### Report the current configuration score

Risk. Ask for the current Microsoft Defender configuration score and the certificate assessments behind it, and get a plain-language summary for the risk committee.

## Elaichi vs Zapier MCP vs Composio for Microsoft Defender

All three can connect Microsoft Defender to an AI assistant, and all three have admin controls. They differ in where access lives and how you pay.

| What to check | Elaichi | Zapier MCP | Composio |
| --- | --- | --- | --- |
| Where the AI connects | One address for the whole organization. Endpoint: https://api.elaichi.ai/mcp | A server per member, created at sign-in. | An MCP endpoint per team, or an SDK. |
| Control over Microsoft Defender tools | Allow or restrict single Microsoft Defender tools, per role or user. | App and action restrictions on the account. | Role permissions, down to the action. |
| Record of calls | One audit entry per Microsoft Defender call. | A History tab of tool calls. | A log of every tool call. |
| Single sign-on | SAML or OIDC, plus SCIM, on Gold. | SAML on Enterprise. | SAML and OIDC on Enterprise. |
| Price | Free for one person; $15 per user per month on Gold. | 2 tasks per successful call. | Billed per tool call. |

Sources: Zapier MCP [docs](https://docs.zapier.com/mcp/get-started/quickstart), [security](https://docs.zapier.com/mcp/manage/security), [usage](https://docs.zapier.com/mcp/features/usage); Composio [docs](https://docs.composio.dev/docs/composio-connect), [gateway](https://composio.dev/mcp-gateway), [enterprise](https://composio.dev/enterprise), [pricing](https://composio.dev/pricing). Checked September 2026.

Longer take: [Zapier MCP alternative](/blog/zapier-mcp-alternative/) and [when you don't need an MCP gateway](/blog/when-you-dont-need-an-mcp-gateway/).

## Frequently asked questions

### How do I connect Microsoft Defender to Claude?

Connect Microsoft Defender in Elaichi first, then in Claude go to Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp. Microsoft Defender needs an OAuth application that your organization registers inside Defender itself, so one administrator brings a client ID and secret once when connecting. After that, everyone signs in through Microsoft's own login screen as themselves and never handles the secret.

### Does Microsoft Defender work with ChatGPT and Cursor as well as Claude?

Yes. Once Microsoft Defender is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, Microsoft Copilot, any other MCP client and the Elaichi Agent. You connect Microsoft Defender once and it is available everywhere.

### What can an AI agent actually do with my Microsoft Defender data?

It can list and read Microsoft Defender alerts, update or create them, and show the machines, users, domains, files and IPs tied to each one. It can pull device antivirus health, baseline compliance, browser extension and certificate assessments, your current configuration score, and export assessments for an audit. Short, concrete asks such as "open alerts on finance laptops this week" work better than long paragraphs.

### Does connecting Microsoft Defender give the AI access to every alert and device?

No. Every call to Microsoft Defender runs as the person who signed in, so an agent sees only the alerts, machines and assessments that person's Defender role already allows. Elaichi can narrow that further with roles and restrictions, and it can never widen it beyond what Microsoft Defender itself grants.

### Can my team share one Microsoft Defender connection?

Yes. One administrator connects Microsoft Defender in Elaichi and shares the connection with a team, and nobody else ever sees the client ID or secret. Each person still signs in to Microsoft Defender as themselves, so the audit log records who looked at which alert.

### Can I stop an agent from changing alerts in Microsoft Defender?

Yes. Restrictions in Elaichi work per action, so you can allow reading Microsoft Defender alerts while restricting updating or creating them, on a role or on one person. A restricted action still shows up in Elaichi's tool search, flagged as restricted, but no client can run it, whatever the prompt says.

### What happens to a Microsoft Defender connection when someone leaves?

Offboarding a person in Elaichi ends their access to Microsoft Defender through every client at once. A shared Microsoft Defender connection keeps working for everyone else on the team. If you want it gone entirely, disconnecting Microsoft Defender once in Elaichi removes it from Claude, ChatGPT, Cursor, Microsoft Copilot and every other client.

### Does the Microsoft Defender MCP connector work with Gemini, Codex, Claude Code or other MCP clients?

Yes. Microsoft Defender is reached over the same MCP endpoint every client uses, so anything that speaks MCP can call it — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor, Microsoft Copilot, and the Elaichi Agent. The tools on offer and the access behind them are identical whichever client asks. Only the setup screen differs.

### Is Elaichi an alternative to Zapier MCP for Microsoft Defender?

Yes. Both let Claude, ChatGPT or Cursor use Microsoft Defender. Zapier MCP fits a team that already automates in Zapier, since each person signs in and acts as themselves in that account. Elaichi fits when IT wants one address for the whole company, per-tool rules by role, and a record of every Microsoft Defender call.

### How is Elaichi different from Composio for Microsoft Defender?

Composio gives AI agents tools and sign-in handling across 1,000+ apps, for developers building agents or people using an assistant, billed per tool call. Elaichi gives a company's own people governed access to Microsoft Defender: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 110 Microsoft Defender tools

Every tool below is callable through https://api.elaichi.ai/mcp once Microsoft Defender is connected, subject to the toolbox it is in and the restrictions on the caller.

- **List all Microsoft Defender alert domains** (List). Retrieves all domains related to a specific alert. Needs delegated permission URL.Read.All. Required: alert_id.
- **List all Microsoft Defender alert files** (List). Retrieves all files related to a specific alert. Needs delegated permission File.Read.All. Required: alert_id.
- **List all Microsoft Defender alert ips** (List). Retrieves all IPs related to a specific alert. Needs delegated permission Ip.Read.All. Required: alert_id.
- **Get single Microsoft Defender alert machine by ID** (Get). Retrieves Device related to a specific alert. Needs delegated permission Machine.Read or Machine.ReadWrite. The id is the alert id. Required: id.
- **Get single Microsoft Defender alert user by ID** (Get). Retrieves the User related to a specific alert. Needs delegated permission User.Read.All. The id is the alert id. Required: id.
- **List all Microsoft Defender alerts** (List). Retrieves a collection of Alerts. Needs delegated permission Alert.ReadWrite.
- **Get single Microsoft Defender alert by ID** (Get). Retrieves specific Alert by its ID. Needs delegated permission Alert.ReadWrite. Required: id.
- **Update a Microsoft Defender alert by ID** (Update). Updates properties of existing Alert. Needs delegated permission Alert.ReadWrite. Required: id.
- **Create a Microsoft Defender alert** (Create). Creates new Alert on top of Event. Needs delegated permission Alert.ReadWrite. Required: eventTime, reportId, machineId, severity, title, description, recommendedAction, category.
- **Microsoft Defender alerts batch update** (Update). Updates properties of a batch of existing Alerts. Needs delegated permission Alert.ReadWrite. Required: alertIds.
- **List all Microsoft Defender baseline compliance assessments** (List). Returns all security baselines assessments for all devices, on a per-device basis. It returns a table with a separate entry for every unique combination of DeviceId, ProfileId, ConfigurationId. Paged JSON response (pageSize up to 200,000). Needs delegated permission SecurityBaselinesAssessment.Read.
- **Microsoft Defender baseline compliance assessments export** (Export). Export the full Security baseline assessment methods and properties per device as downloadable files. Returns SAS URLs of gzip-compressed JSON-lines files (for large tenants); download them with any HTTP client. Needs delegated permission SecurityBaselinesAssessment.Read.
- **List all Microsoft Defender baseline configurations** (List). This API retrieves a list of the configurations being assessed in active baseline profiles.
- **List all Microsoft Defender baseline profiles** (List). Provides information about the security baselines assessment profiles APIs that pull "Microsoft Defender Vulnerability Management" data. There are different API calls to get different types of data.
- **List all Microsoft Defender browser extension assessments** (List). This API response contains all the data for installed browser extensions per device. Returns a table with an entry for every unique combination of DeviceId, BrowserName, ExtensionId. Paged JSON response (pageSize up to 200,000). Needs delegated permission Software.Read.
- **Microsoft Defender browser extension assessments export** (Export). Export the full browser extensions assessment as downloadable files. Returns SAS URLs of gzip-compressed JSON-lines files (for large tenants); download them with any HTTP client. Needs delegated permission Software.Read.
- **List all Microsoft Defender browser extension permissions** (List). Retrieves a list of all the permissions requested by a specific browser extension. This is a static data description and would mainly be used to enhance the data returned by the Export browser extensions assessment API. Needs delegated permission Software.Read.
- **List all Microsoft Defender certificate assessments** (List). Returns all certificate assessments for all devices, on a per-device basis. It returns a table with a separate entry for every unique combination of DeviceId, Thumbprint and Path. Paged JSON response (pageSize up to 200,000). Needs delegated permission Vulnerability.Read.
- **Microsoft Defender certificate assessments export** (Export). Export the full Certificate assessment methods and properties per device as downloadable files. Returns SAS URLs of gzip-compressed JSON-lines files (for large tenants); download them with any HTTP client. Needs delegated permission Vulnerability.Read.
- **Microsoft Defender configuration score get current** (Get). Retrieves the organizational device secure score. Needs delegated permission Score.Read.
- **List all Microsoft Defender device av health** (List). This API retrieves a list of Microsoft Defender Antivirus device antivirus health details. Returns a table with an entry for every unique combination of: Needs delegated permission Machine.Read or Vulnerability.Read.
- **List all Microsoft Defender domain alerts** (List). Retrieves a collection of Alerts related to a given domain address. Needs delegated permission Alert.ReadWrite. Required: domain.
- **List all Microsoft Defender domain machines** (List). Retrieves a collection of Machines that have communicated to or from a given domain address. Needs delegated permission Machine.ReadWrite. Required: domain.
- **List all Microsoft Defender domain stats** (List). Retrieves the statistics on the given domain. Needs delegated permission URL.Read.All. Required: domain.
- **Microsoft Defender exposure score get current** (Get). Retrieves the organizational exposure score. Needs delegated permission Score.Read.
- **List all Microsoft Defender exposure score by machine groups** (List). Retrieves a list of exposure scores by device group. Needs delegated permission Score.Read.
- **List all Microsoft Defender file alerts** (List). Retrieves a collection of alerts related to a given file hash. Needs delegated permission Alert.ReadWrite. Required: file_hash.
- **List all Microsoft Defender file machines** (List). Retrieves a collection of Machines related to a given file hash. Needs delegated permission Machine.ReadWrite. Required: file_hash.
- **List all Microsoft Defender file stats** (List). Retrieves the statistics for the given file. Needs delegated permission File.Read.All. Required: file_hash.
- **Get single Microsoft Defender file by ID** (Get). Retrieves a File by identifier Sha1, or Sha256 Needs delegated permission File.Read.All. The id is the file SHA1 or SHA256 hash. Required: id.
- **List all Microsoft Defender indicators** (List). Retrieves a collection of all active Indicators. Needs delegated permission Ti.ReadWrite.
- **Create a Microsoft Defender indicator** (Create). Submits or Updates new Indicator entity. Needs delegated permission Ti.ReadWrite. Required: indicatorValue, indicatorType, action, title, description.
- **Delete a Microsoft Defender indicator by ID** (Delete). Deletes an Indicator entity by ID. Required: id.
- **Microsoft Defender indicators import** (Import). Submits or Updates batch of Indicator entities. Needs delegated permission Ti.ReadWrite. Required: Indicators.
- **Microsoft Defender indicators batch delete** (Delete). Deletes Indicator entities by ID. Required: IndicatorIds.
- **Microsoft Defender info gathering assessments export** (Export). Export the full information gathering assessment as downloadable files. Returns SAS URLs of gzip-compressed JSON-lines files (for large tenants); download them with any HTTP client. Needs delegated permission Vulnerability.Read.
- **List all Microsoft Defender investigations** (List). Retrieves a collection of Investigations. Needs delegated permission Alert.ReadWrite.
- **Get single Microsoft Defender investigation by ID** (Get). Retrieves specific Investigation by its ID. ID can be the investigation ID or the investigation triggering alert ID. Needs delegated permission Alert.ReadWrite. Required: id.
- **List all Microsoft Defender IP alerts** (List). Retrieves a collection of alerts related to a given IP address. Needs delegated permission Alert.Read or Alert.ReadWrite. Required: ip.
- **List all Microsoft Defender IP stats** (List). Retrieves the statistics for the given IP. This statistic information is based on data from the past 30 days. Needs delegated permission Ip.Read.All. Required: ip.
- **List all Microsoft Defender library files** (List). List live response library files. Needs delegated permission Library.Manage.
- **Create a Microsoft Defender library file** (Create). Upload file to live response library. Needs delegated permission Library.Manage. Required: File.
- **Delete a Microsoft Defender library file by ID** (Delete). Delete a file from live response library. Needs delegated permission Library.Manage. The id is the file name. Required: id.
- **List all Microsoft Defender machine actions** (List). Retrieves a collection of Machine Actions. Needs delegated permission Machine.ReadWrite.
- **Get single Microsoft Defender machine action by ID** (Get). Retrieves specific Machine Action by its ID. Needs delegated permission Machine.ReadWrite. Required: id.
- **Microsoft Defender machine actions cancel** (Cancel). Cancel an already launched machine action that isn't yet in final state (completed, canceled, failed). Needs delegated permission Machine.CollectForensics Machine.Isolate Machine.RestrictExecution Machine.Scan Machine.Offboard Machine.StopAndQuarantineMachine.LiveResponse. Required: machine_action_id.
- **Microsoft Defender machine actions get package uri** (Get). Get a URI that allows downloading of an Investigation package. Needs delegated permission Machine.CollectForensics. Required: machine_action_id.
- **Microsoft Defender machine actions get live response result link** (Get). Retrieves a specific live response command result by its index. Needs delegated permission Machine.LiveResponse. Required: machine_action_id, index.
- **List all Microsoft Defender machine alerts** (List). Retrieves all Alerts related to a specific device. Needs delegated permission Alert.ReadWrite. Required: machine_id.
- **List all Microsoft Defender machine logon users** (List). Retrieves a collection of logged on users on a specific device. Needs delegated permission User.Read.All. Required: machine_id.
- ...and 60 more tools. Call `tools/list` via the MCP endpoint, or see the full catalog via the API, for the complete set.
