# SentinelOne MCP connector

Connecting SentinelOne to Elaichi lets Claude, ChatGPT, Cursor, Microsoft Copilot and the Elaichi Agent look up accounts, agents and activities, start scans and disconnect endpoints, each person signing in as themselves with every action logged.

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

## Facts

| | |
| --- | --- |
| Application | SentinelOne |
| Category | Security |
| AI tools | 739 |
| 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 SentinelOne is connected

- Which SentinelOne agents have been disconnected from the network this week?
- Start a full disk scan on the finance team's laptops.
- Summarize yesterday's activities across all SentinelOne accounts.

## Connect SentinelOne in Elaichi

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

1. Open Connections, choose Add connection, and pick SentinelOne.
2. Optionally set Share with, then press Connect.
3. Paste a SentinelOne API key. One person generates a token in SentinelOne 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 SentinelOne tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## SentinelOne 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.

## SentinelOne 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`.

## SentinelOne 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.

## SentinelOne 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.

## SentinelOne 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 SentinelOne 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 SentinelOne 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 SentinelOne through Elaichi

### Isolate a suspicious endpoint during triage

Security operations. Ask which agents have shown unusual activity today, then disconnect the one that worries you from the network while you investigate. The disconnect is logged against your name.

### Start a scan without opening the console

Security operations. Kick off a full scan on a laptop, a department, or every agent in an account, and abort it if it was the wrong target.

### Pull logs from an agent for a ticket

IT. When a user reports a slow machine, fetch the agent log straight from the chat and attach what you find to the ticket.

### Retire machines cleanly at refresh time

IT. Decommission or uninstall the agent on laptops that are leaving the fleet, and generate an uninstall password only for the ones that need it.

### Export activities for the quarterly review

Compliance. Ask for every activity across your SentinelOne accounts in a date range, grouped by type, and export it for the auditors.

### Keep client accounts current

Partner management. List all SentinelOne accounts you manage, update the ones whose details changed, and reactivate or expire accounts in bulk when a contract starts or ends.

## Elaichi vs Zapier MCP vs Composio for SentinelOne

All three can connect SentinelOne 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 SentinelOne tools | Allow or restrict single SentinelOne 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 SentinelOne 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 SentinelOne to Claude?

Connect SentinelOne in Elaichi first by pasting an API key from your SentinelOne console. 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 https://api.elaichi.ai/mcp. Claude signs you in through Elaichi and SentinelOne is ready to use.

### Does SentinelOne work with ChatGPT and Cursor as well as Claude?

Yes. Once SentinelOne 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 SentinelOne once and every client picks it up.

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

It can list your SentinelOne accounts and agents, show recent activities and export them, start or abort a scan on an endpoint, fetch an agent's log, and disconnect, reconnect, decommission or uninstall an agent. It can also create and update accounts and manage uninstall passwords. Short, concrete asks such as "disconnect the agent on Priya's laptop" work better than long sentences.

### Does connecting SentinelOne give the AI access to every account and agent?

No. Every call to SentinelOne runs inside the access of the person who signed in, so an agent sees only the accounts and endpoints that person can already see in SentinelOne. Elaichi can narrow that further with roles and restrictions, and it never widens it.

### Can my team share one SentinelOne connection?

Yes. One person connects SentinelOne with the API key and shares the connection with a team in Elaichi, and nobody else on the team ever handles the key. Each person still signs in to Elaichi as themselves, so the audit log names who ran each scan or disconnect.

### Can I stop an agent from deleting or changing things in SentinelOne?

Yes. Restrictions in Elaichi work per action, so you can allow looking up SentinelOne agents and activities while restricting decommission, uninstall or account updates, on a role or on one person. A restricted action still appears in Elaichi's tool search, flagged as restricted, but no client can run it, whatever the prompt says.

### What happens to a SentinelOne connection when someone leaves?

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

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

Yes. SentinelOne 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 SentinelOne?

Yes. Both let Claude, ChatGPT or Cursor use SentinelOne. 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 SentinelOne call.

### How is Elaichi different from Composio for SentinelOne?

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 SentinelOne: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 739 SentinelOne tools

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

- **List all SentinelOne accounts** (List). Get Accounts. Get the Accounts, and their data, that match the filter. This command gives the Account IDs, which other commands require. Accounts are created by a Global User or by SentinelOne. Each Account contains Sites, which can inherit assets and settings. Permission: Accounts.view.
- **Create a SentinelOne account** (Create). Create Account. Create a new Account. This command requires Global permissions and an MSSP deployment. Consult with your SE before you run this command. An Account is a logical segment with permissions to configure features for specific Sites. Required: data. Permission: Accounts.create.
- **Get single SentinelOne account by ID** (Get). Get Account by ID. Get Account data from a given Account ID. To get an Account ID, run "accounts". Required: id. Permission: Accounts.view.
- **Update a SentinelOne account by ID** (Update). Update Account. Change the data of an Account. This command requires a Global user or an Account user and Admin role. Use this command to change the name, ID, SKUs and how they are distributed among Sites and Agents, and more. Required: id, data. Permission: Accounts.edit.
- **SentinelOne accounts revert policy bulk update** (Update). Revert Policy. The policy of the Account is based on the default Global policy and is enforced by all endpoints in the Sites and Groups of the Account (if you did not change the Site or Group policies). Required: account_id. Permission: Policy.edit.
- **SentinelOne accounts reactivate bulk update** (Reactivate). Reactivate Account. Reactivate an expired Account. This command requires a Global user or Support. Consult with your SentinelOne SE. Required: account_id, data. Permission: Accounts.edit.
- **Create a SentinelOne accounts expire now** (Create). Expire an Account. Expire an Account immediately. The user must have Global access or Account acces with permissions for the Account. Best practice: Consult with Support before you use this command. Required: account_id. Permission: Accounts.edit.
- **List all SentinelOne accounts uninstall password metadata** (List). Get Uninstall Password Metadata. Get the uninstall password metadata, such as which user created and revoked it and when. Required: account_id. Permission: Accounts.view.
- **List all SentinelOne accounts uninstall password view** (List). Get Uninstall Password. Get the uninstall password to uninstall several Agents of one Account with one command. Required: account_id. Permission: Endpoints.viewUninstallPassword.
- **Create a SentinelOne accounts uninstall password generate** (Create). Generate/Regenerate Uninstall Password. You can uninstall all Agents of one Account with one command that requires a password. This command sets a new account-level uninstall password. To enable this feature, submit a ticket with Support. Required: account_id, data. Permission: Endpoints.modifyUninstallPassword.
- **Create a SentinelOne accounts uninstall password revoke** (Create). Revoke Uninstall Password. Delete the account-level uninstall password. If you do not delete it, you or another Console user can mistakenly use the Account passphrase (and uninstall all Agents) when you mean to uninstall one Agent. Required: account_id. Permission: Endpoints.modifyUninstallPassword.
- **List all SentinelOne export accounts** (List). Export Accounts. Export Accounts data to a CSV, for Accounts that match the filter. Permission: Accounts.view.
- **List all SentinelOne activities** (List). Get Activities. Get the activities, and their data, that match the filters. We recommend that you set some values for the filters. The full list will be too large to be useful. Permission: Activity Page.view.
- **List all SentinelOne activities types** (List). Get Activity Types. Get a list of activity types. This is useful to see valid values to filter activities in other commands. Permission: Activity Page.view.
- **List all SentinelOne last activity as syslog** (List). Last activity as Syslog message. To see examples of Syslog messages, you can get the Syslog message that corresponds to the last activity that matches the filter. This is not intended for production purposes. Permission: Activity Page.view.
- **List all SentinelOne export activities** (List). Export Activities. Export the list of activities. Permission: Activity Page.view.
- **Create a SentinelOne agents actions broadcast** (Create). Broadcast Message. You can send a message through the Agents that users can see. This is useful for endpoints that have human users. This command is supported on Windows and macOS endpoints (not supported on Linux). The message is sent to all endpoints that match the filter. Required: data, filter. Permission: Endpoints.sendMessage.
- **Create a SentinelOne agents actions connect** (Create). Connect to Network. After you run "disconnect from network" on endpoints, analyze the issue, and mitigate threats. Use this command to reconnect to the network all endpoints that match the filter. To learn more, see "Disconnect from Network". Required: filter. Permission: Endpoints.reconnectToNetwork.
- **Create a SentinelOne agents actions fetch log** (Create). Fetch Logs. Get the Agent and Endpoint logs from Agents that match the filter. The Agent logs are encrypted and only Support can read them. The Endpoint logs, for operations on the computers, laptops, or servers that have the Agent installed, are readable. Required: filter. Permission: Endpoints.FetchLogs.
- **Create a SentinelOne agents actions initiate scan** (Create). Initiate Scan. Use this command to run a Full Disk Scan on Agents that match the filter. Full Disk Scan finds dormant suspicious activity, threats, and compliance violations, that are then mitigated according to the policy. It scans the local file system. Required: filter. Permission: Endpoints.initiateScan.
- **Create a SentinelOne agents actions abort scan** (Create). Abort Scan. Immediately stop a Full Disk Scan on all Agents that match the filter. See "Initiate scan" to learn more about Full Disk Scan. Required: filter. Permission: Endpoints.abortScan.
- **Create a SentinelOne agents actions disconnect** (Create). Disconnect from Network. Use this command to isolate (quarantine) endpoints from the network, if the endpoints match the filter. The Agent can communicate with the Management, which lets you analyze and mitigate threats. Required: filter. Permission: Endpoints.disconnectFromNetwork.
- **Create a SentinelOne agents actions decommission** (Create). Decommission. If a user is scheduled for time off, or a device is scheduled for maintenance, you can decommission the Agent. This removes the Agent from the Management Console. Required: filter. Permission: Endpoints.decommission.
- **Create a SentinelOne agents actions uninstall** (Create). Uninstall. Use this command to uninstall Agents that match the filter. For Windows and macOS, make sure that all remnants of the Agent are removed: reboot the endpoints after uninstall. Use the "restart" command. Required: filter. Permission: Endpoints.uninstall.
- **Create a SentinelOne agents actions restart machine** (Create). Restart. Use this command to restart endpoints that have an Agent installed and that fit the filter. We recommend that you use the "broadcast" command to send a message to users of endpoints before you restart their computers. Required: filter. Permission: Endpoints.reboot.
- **Create a SentinelOne agents actions shutdown** (Create). Shutdown. You can shut down endpoints remotely for performance, maintenance, or security. This command shuts down all endpoints that match the filter. Best Practice: If an endpoint is infected, we recommend the "disconnect" command and not the "shutdown" command. Required: filter. Permission: Endpoints.shutDown.
- **Create a SentinelOne agents actions approve uninstall** (Create). Approve Uninstall. If a user tries to uninstall the SentinelOne Agent from an endpoint, an uninstall request is sent to the Management. You must approve the request. After you approve a request, users see a message that the request was approved. Required: filter. Permission: Endpoints.approveUninstall.
- **Create a SentinelOne agents actions reject uninstall** (Create). Reject uninstall. Reject uninstall requests for all Agents that match the filter. To learn more about Uninstall Requests, see "Approve Uninstall". Required: filter. Permission: Endpoints.rejectUninstall.
- **Create a SentinelOne agents actions update software** (Create). Update Software. Use this command to update the Agent version on endpoints that have the Agent installed and that match the filter. For a cloud-based Management, SentinelOne updates your Management Console with the latest Agent versions. Required: data, filter. Permission: Endpoints.updateSoftware.
- **Create a SentinelOne agents actions reset local config** (Create). Reset Local Config. SentinelCtl is the CLI for Agents. It runs commands directly on one Agent at a time. You can use this command to clear the SentinelCtl changes from all Agents that match the filter. Required: filter. Permission: Endpoints.resetLocalConfiguration.
- **Create a SentinelOne agents actions set config** (Create). Set Persistent Configuration Overrides. This command requires Global permissions or Support. The configuration of an Agent can be changed in different ways, such as through Policy settings, Policy Override, SentinelCtl, and changes to the LocalConfig.json file. Required: data, filter. Permission: Endpoints.configuration.
- **Create a SentinelOne agents actions set external ID** (Create). Set External ID. You can add a Customer Identifier (a string) to identify each endpoint or to tag sets of endpoints. The string shows in the Endpoint Details of the Management Console. Required: data, filter. Permission: Endpoints.setCustomerIdentifier.
- **Create a SentinelOne agents actions fetch file** (Create). Fetch Files. Fetch files from endpoints (up to 10 MB for each command) to analyze the root of threats (that come from files - of course, this does not help for fileless threats). Set the pathnames in the body of the request. Required: agent_id, data. Permission: Endpoints.fileFetch.
- **Create a SentinelOne agents actions move to site** (Create). Move between Sites. This command requires Account or Global level access. Agents are assigned to a Site when they are first installed with a Site Token. Required: data, filter. Permission: Endpoints.moveToAnotherSite.
- **Create a SentinelOne agents actions fetch firewall rule** (Create). Fetch Firewall Rules. Firewall Control is disabled at the Global level. When it is first enabled, all Sites and Groups inherit the Firewall Control policy from the Global policy. Required: data, filter. Permission: Endpoints.configureFirewallLogging.
- **Create a SentinelOne agents actions move to console** (Create). Move to Console. You can move Agents between Management Consoles. This command moves Agents to a target Console, Account, and Site, given the Console URL and Site token. You must have Global permissions for the source Console and access to the Site token of the target Site. Required: data, filter. Permission: Endpoints.migrateAgent.
- **Create a SentinelOne agents actions fetch installed app** (Create). Get Applications. Application Risk Management is an EA feature. Contact your partner or SentinelOne SE to learn how to join the EA program. Required: filter. Permission: Endpoints.showApplications.
- **Create a SentinelOne agents actions start remote shell** (Create). Start Remote Shell. Remote shell is an opened websocket between the browser and the Agent, with a proprietary communication protocol that requires an unreasonable effort to run from the API. We recommend that you not use this call. Required: data, filter. Permission: Endpoints.remoteShell.
- **Create a SentinelOne agents actions can start remote shell** (Create). Can run Remote Shell. Who can run Remote Shell? Remote Shell is a powerful way to respond remotely to events on endpoints. It lets you open full shell capabilities - PowerShell on Windows and Bash on macOS and Linux. Required: filter. Permission: Endpoints.remoteShell.
- **Create a SentinelOne agents actions terminate remote shell** (Create). Terminate Remote Shell. Remote Shell is a powerful, full shell for Windows, macOS, and Linux. It is best practice to terminate Remote Shell sessions when they are not in use. Required: data, filter. Permission: Endpoints.remoteShell.
- **Create a SentinelOne agents actions firewall logging** (Create). Fetch Firewall Logs. Get Firewall Control events in the local log file, written in clear text, for Firewall Control events of an endpoint with Firewall Control enabled. Enable the logs for Agents that match the filter. Required: data, filter. Permission: Endpoints.configureFirewallLogging.
- **Create a SentinelOne agents actions randomize uuid** (Create). Randomize UUID. IMPORTANT: This action will assign a new UUID to Agents that match the filter. Run it only when instructed to do so by SentinelOne Support. If you clone the Agent on a VM or VDI without the /VDI switch, you might need to run this command. Required: filter. Permission: Endpoints.randomizeUuid.
- **Create a SentinelOne agents actions mark up to date** (Create). Mark as up-to-date. The value of the Agent version as "up-to-date" is a useful filter for many actions. There are scenarios where the Management does not recognize a version as latest. Required: filter. Permission: Endpoints.markAsUpToDate.
- **Create a SentinelOne agents actions ranger enable** (Create). Enable Network Discovery. Singularity Network Discovery gives full visibility of all devices connected to your network. Network Discovery scans your corporate environment to identify and manage connected devices, even those not protected by or supported by SentinelOne. Required: filter. Permission: Endpoints.enableRanger.
- **Create a SentinelOne agents actions ranger disable** (Create). Disable Network Discovery. Disable Network Discovery from the Agents that match the filter. Singularity Network Discovery gives full visibility of all devices connected to your network. Required: filter. Permission: Endpoints.disableRanger.
- **Create a SentinelOne agents actions local upgrade authorization** (Create). Edit local upgrade/downgrade Site authorization. Edit when authorization of local upgrades/downgrades expires. Required: data, filter. Permission: Local Upgrade/Downgrade Authorization.edit.
- **Create a SentinelOne agents actions disable agent** (Create). Disable Agent. Use this command to disable Agents that match the filter. Disabled agents run with minimal footprint and do not detect or mitigate threats, but they maintain connectivity with the Management Console. Required: data, filter. Permission: Endpoints.disableAgent.
- **Create a SentinelOne agents actions enable agent** (Create). Enable Agent. Use this command to enable disabled Agents that match the filter. If the command returns "Insufficient permissions", make sure you have permissions for the Account, Site, or Group and a role that allows Disable Agent (Admin, IR team or IT). Required: data, filter. Permission: Endpoints.enableAgent.
- **Create a SentinelOne agents actions start profiling** (Create). Start Remote Profiling. Use this command to start remote profiling on Agents that match the filter. Remote profiling lets you collect runtime diagnostic information for Agents on containers. Required: filter. Permission: Endpoints.remoteProfiling.
- **Create a SentinelOne agents actions stop profiling** (Create). Stop Remote Profiling. Use this command to stop remote profiling on Agents that match the filter. If the command returns "Insufficient permissions", make sure you have permissions for the Account, Site, or Group and a role that allows Stop Remote Profiling (Admin or IT). Required: filter. Permission: Endpoints.remoteProfiling.
- ...and 689 more tools. Call `tools/list` via the MCP endpoint, or see the full catalog via the API, for the complete set.
