# Iru (formerly Kandji) MCP connector

The Iru (formerly Kandji) connector brings your Apple device fleet, installed apps, library items, blueprints and users into Claude, ChatGPT, Cursor, Microsoft Copilot and any MCP client, so your team can ask about and update devices while every call stays inside their own Iru access.

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

## Facts

| | |
| --- | --- |
| Application | Iru (formerly Kandji) |
| Category | Device Management |
| AI tools | 17 |
| Authentication | Connects with an API key |
| Bring your own OAuth app | No |
| 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 Iru (formerly Kandji) is connected

- List all Iru devices missing their assigned blueprint.
- Which Iru devices have apps out of date?
- Show Iru users with more than two devices.

## Connect Iru (formerly Kandji) in Elaichi

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

1. Open Connections, choose Add connection, and pick Iru (formerly Kandji).
2. Optionally set Share with, then press Connect.
3. Paste an Iru (formerly Kandji) API key. One person generates a token in Iru (formerly Kandji) and pastes it once. Everyone else works through Share with, and never sees it.

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

## Iru (formerly Kandji) 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.

## Iru (formerly Kandji) 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`.

## Iru (formerly Kandji) 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.

## Iru (formerly Kandji) 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.

## Iru (formerly Kandji) 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 Iru (formerly Kandji) 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 data and remove access", which you tick yourself. Calling a Iru (formerly Kandji) tool needs "Run your connected tools". 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 Run your connected tools, which can also delete in connected apps and cannot be undone.

## What teams do with Iru (formerly Kandji) through Elaichi

### Find out who has which Mac

IT. Ask which devices are enrolled in Iru (formerly Kandji), who each one is assigned to, and when it last checked in, without opening the console or exporting a spreadsheet.

### Check installed apps before replying to a ticket

Help desk. Pull up a colleague's device in Iru (formerly Kandji), see which apps are installed and which library items have reached it, and answer the ticket with facts instead of a guess.

### Confirm required software reached every device

Security. Compare the apps and library items on each Iru (formerly Kandji) device against what should be there, and get a list of the machines that are missing something.

### Understand what a blueprint actually sets

IT. Look at the blueprints in Iru (formerly Kandji), read what a specific one configures, and see which devices sit under it before changing a policy.

### Clean up when someone leaves

People. Look up the departing person's user record and devices in Iru (formerly Kandji), update the device details, and remove the user record once hardware is returned.

### Keep the hardware inventory current

Operations. Update asset tags and assigned users on Iru (formerly Kandji) devices as laptops move between people and offices, so the device list matches what is on desks.

## Elaichi vs Zapier MCP vs Composio for Iru (formerly Kandji)

All three can connect Iru (formerly Kandji) 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 Iru (formerly Kandji) tools | Allow or restrict single Iru (formerly Kandji) 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 Iru (formerly Kandji) 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 Iru (formerly Kandji) to Claude?

Connect Iru (formerly Kandji) in Elaichi first, which asks you to paste an API key from your Iru settings. There is no OAuth application to register and no client ID or secret to generate. Then open Claude, go to Customize, then Connectors, then Add, and paste the endpoint https://api.elaichi.ai/mcp. Claude will ask you to sign in to Elaichi as yourself, and your Iru devices, users and blueprints are available from then on.

### Does Iru (formerly Kandji) work with ChatGPT and Cursor as well as Claude?

Yes. Once Iru (formerly Kandji) 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 Iru once and every client you use picks it up.

### What can an AI agent actually do with my Iru (formerly Kandji) data?

An agent connected to Iru (formerly Kandji) can list your enrolled devices, open a single device and report who it is assigned to and when it last checked in, and show which apps and library items are on it. It can read your blueprints, look up users, update device details such as asset tag or assigned user, and remove a device or user record when you ask it to. It cannot do anything in Iru that you could not do yourself.

### Does connecting Iru (formerly Kandji) give the AI access to my whole device fleet?

No. Every call to Iru (formerly Kandji) runs inside the access of the person who signed in, so the AI sees the devices, users and blueprints that person's Iru permissions allow and nothing more. Elaichi can narrow that further by restricting actions like deleting devices, but it can never grant more than the person already has in Iru.

### Can my team share one Iru (formerly Kandji) connection?

Yes. One administrator connects Iru (formerly Kandji) with the API key and shares the connection with a team in Elaichi, and nobody else ever sees or handles that key. Each teammate still signs in to Elaichi as themselves, so the audit log names the actual person who looked up a device or updated a user, not a shared account.

### Can I stop an agent from deleting or changing things in Iru (formerly Kandji)?

Yes. Restrictions in Elaichi apply per action, so you can allow reading Iru (formerly Kandji) devices, apps and blueprints while restricting updates and deletions of devices or users. A restricted action still appears in Elaichi's tool search, flagged as restricted, but Claude, ChatGPT or any other client cannot run it, whatever the prompt says.

### What happens to an Iru (formerly Kandji) connection when someone leaves?

When you offboard a person in Elaichi, their access to Iru (formerly Kandji) through every AI client ends at once. If they had shared an Iru connection with a team, it keeps working for everyone else. If you want Iru gone entirely, disconnecting it once in Elaichi removes it from Claude, ChatGPT, Cursor, Microsoft Copilot and every other client at the same time.

### Does the Iru (formerly Kandji) MCP connector work with Gemini, Codex, Claude Code or other MCP clients?

Yes. Iru (formerly Kandji) 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 Iru (formerly Kandji)?

Yes. Both let Claude, ChatGPT or Cursor use Iru (formerly Kandji). 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 Iru (formerly Kandji) call.

### How is Elaichi different from Composio for Iru (formerly Kandji)?

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 Iru (formerly Kandji): one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 17 Iru (formerly Kandji) tools

Every tool below is callable through https://api.elaichi.ai/mcp once Iru (formerly Kandji) is connected, subject to the toolbox it is in and the restrictions on the caller.

- **List all Iru (formerly Kandji) devices** (List). Get a list of devices in Kandji. Returns fields including device_id, device_name, model, platform, os_version, serial_number, user, and tags.
- **Get single Iru (formerly Kandji) device by ID** (Get). Get device details for a specified device in Kandji. Requires device id. Returns device information including hardware, software, and status fields.
- **Update a Iru (formerly Kandji) device by ID** (Update). Update device information in Kandji using id. Supports updating user assignment, asset_tag, blueprint_id, and tags. Use null to clear asset_tag or user, and empty list to clear tags. Returns updated device fields.
- **Delete a Iru (formerly Kandji) device by ID** (Delete). Delete a specific device in Kandji using id. This removes the device record, unenrolls it from MDM, and automatically uninstalls the agent on next check-in for macOS and Windows devices. Returns no content.
- **List all Iru (formerly Kandji) device apps** (List). Get a list of all installed apps for a specified device in Kandji. Requires device_id. The response includes app details, such as name and version.
- **List all Iru (formerly Kandji) device library items** (List). Get all library items and their statuses for a specified device in Kandji. Returns fields including library item status indicating availability, installation state, and compatibility.
- **List all Iru (formerly Kandji) blueprints** (List). Get a list of blueprints in Kandji. Returns blueprint records with details such as id and name.
- **Get single Iru (formerly Kandji) blueprint by ID** (Get). Get information about a specific blueprint in Kandji using id. Returns blueprint details including configuration and metadata.
- **List all Iru (formerly Kandji) users** (List). List users in Kandji. Returns an array of users with key details such as id, email, name, active, archived, created_at, updated_at, department, job_title, device_count, and integration information.
- **Get single Iru (formerly Kandji) user by ID** (Get). Get details for a specific user in Kandji using id. Returns key fields including name, email, active status, department, job_title, device_count, and integration details such as id, name, and type.
- **Delete a Iru (formerly Kandji) user by ID** (Delete). Delete a specific user in Kandji by id. Returns no content on success. If the user is still assigned to one or more devices, a 400 error with 'detail' explaining the assignment issue is returned.
- **Get single Iru (formerly Kandji) device detail by ID** (Get). Get the full details of a specific device in Iru (Kandji). Requires device id. Returns sections tailored to the device family, such as general, hardware_overview, network, mdm, filevault, kandji_agent, security_information, activation_lock, lost_mode, management, and security.
- **Get single Iru (formerly Kandji) device summary by ID** (Get). Get the summary of a specific device in Iru (Kandji). Requires device id. Returns fields including device_id, device_name, model, platform, os_version, serial_number, user, blueprint_id, mdm_enabled, agent_installed, last_check_in, and tags.
- **List all Iru (formerly Kandji) prism filevault** (List). Get FileVault information for macOS devices in Iru (Kandji) from Prism. Returns one row per device with fields including device_id, device__name, serial_number, status, key_escrowed, and last_collected_at. Use filter to narrow the rows, for example by device_id.
- **List all Iru (formerly Kandji) prism application firewall** (List). Get Application Firewall details for macOS devices in Iru (Kandji) from Prism. Returns one row per device with fields including device_id, device__name, serial_number, status, block_all_incoming, stealth_mode, logging, and last_collected_at. Use filter to narrow the rows, for example by device_id.
- **List all Iru (formerly Kandji) prism device information** (List). Get attributes about devices in Iru (Kandji) from Prism. Returns one row per device with fields including device_id, device__name, device__family, serial_number, model, model_name, os_version, os_build, host_name, local_hostname, mdm_enabled, agent_installed, first_enrollment_date, last_enrollment_date, and last_checkin. Use filter to narrow the rows, for example by device_id.
- **List all Iru (formerly Kandji) prism apps** (List). Get the applications installed on macOS, iOS, iPadOS, and tvOS devices in Iru (Kandji) from Prism. Returns one row per application per device with fields including device_id, device__name, device__family, serial_number, name, bundle_id, version, path, obtained_from, developer_name, and last_collected_at. Use filter to narrow the rows, for example by device_id or name.
