ESET PROTECT MCP Server
This MCP server provides read-only ESET PROTECT fleet and security visibility for AI assistants and the WYRE Conduit gateway.
List device groups and group members, list/filter devices, and get or batch-get devices by UUID.
Investigate detections via v1 and v2 APIs, detection groups, EDR rules, and EDR rule exclusions.
Review OS/device vulnerabilities, recent vulnerability scans, and vulnerable devices.
Inspect patch status, recent application-patching details, and patching-process history.
List/filter ESET Business Account users and get or batch-get users by UUID.
Use pagination, time-window filters, status/type filters, and batch retrieval where supported.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@ESET PROTECT MCP Server show me devices with known vulnerabilities"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
ESET PROTECT MCP Server
MCP server for ESET PROTECT's ESET Connect API gateway - endpoint fleet visibility (devices and groups), detection/EDR investigation, vulnerability data, patch status, and user management - for AI assistants and the WYRE Conduit gateway.
Authentication
Create a dedicated API user in ESET Business Account: User Management > create a user with read-only permission > enable the Integrations toggle. This gives you an email/password pair for that user.
ESET Connect's token endpoint (POST /oauth/token) is documented in its own OpenAPI spec as a password grant - username/password fields, not client_id/client_secret with grant_type=client_credentials. This server accepts clientId/clientSecret (matching the credential shape used across WYRE's other MCP connectors) and sends them as username/password in that grant. It handles the resulting bearer token's caching and refresh (via the returned refresh_token, falling back to a full re-login if the refresh token itself is rejected) internally - callers only ever supply the long-lived email/password pair.
Related MCP server: XFA MCP
Configuration
Env var | Description |
| Email of the dedicated ESET Connect API user. |
| Password of that API user. |
|
|
|
|
|
|
| When set, the HTTP transport requires a valid |
|
|
Each ESET Connect API domain (device, incident, vulnerability, patch, and user management) is hosted on its own per-region subdomain, e.g. https://us.device-management.eset.systems; the auth host follows the same regional pattern, e.g. https://us.business-account.iam.eset.systems. ESETPROTECT_REGION selects the prefix for all of them.
Tools
Device Management
esetprotect_list_device_groups- list device groups.esetprotect_list_group_devices- list the devices in a group.esetprotect_list_devices- list managed devices, filterable by display name, functionality status, or mute state.esetprotect_get_device- get a device by UUID.esetprotect_batch_get_devices- get multiple devices by UUID in one call.
Incident Management (detections, detection groups, EDR)
esetprotect_list_detections/esetprotect_get_detection- v1 detections (legacy surface).esetprotect_list_detections_v2/esetprotect_get_detection_v2/esetprotect_batch_get_detections- v2 detections (endpoint, cloud office, and EDR).esetprotect_list_detection_groups/esetprotect_get_detection_group- detections clustered into groups.esetprotect_list_edr_rules/esetprotect_get_edr_rule- EDR detection rules.esetprotect_list_edr_rule_exclusions/esetprotect_get_edr_rule_exclusion- EDR rule exclusions.
Vulnerability Management
esetprotect_list_device_os_vulnerabilities- OS vulnerabilities found on devices.esetprotect_list_device_vulnerabilities- vulnerabilities found on devices (application/OS/package scope).esetprotect_list_recent_scans- recent vulnerability scan details.esetprotect_list_vulnerable_devices- devices with known vulnerabilities.
Patch Management
esetprotect_list_recent_application_patching_details- most recent application-patching processes fleet-wide.esetprotect_list_device_patches- patches for devices, filterable by patch type.esetprotect_list_patching_process_details- detailed patching-process history within a time window.
User Management
esetprotect_list_users- list users, filterable by email, display name, protection status, and more.esetprotect_get_user- get a user by UUID.esetprotect_batch_get_users- get multiple users by UUID in one call (atomic - all or none).
Request limits
HTTP request bodies are capped at 256 KiB (262144 bytes). After the
X-Gateway-S2Scheck passes, aContent-Lengthabove the cap returns 413 with a JSON-RPC error; bodies withoutContent-Lengthare capped by the MCP SDK transport. Unauthenticated requests still get 401 first.tools/callarguments are capped at 10,000 elements (every array item and object key at any depth). Larger inputs return anisErrortool result and are never dispatched.
These limits address CVE-2026-111768 / GHSA-46jr-4wx8-v453 (memory exhaustion during validation) and GHSA-f897-62hf-6gxq (no request body / batch cap).
Scope
This is a v1, read-only fleet/security-visibility surface. Explicitly deferred: device move/rename and batch import, detection and detection-group resolve, EDR rule and rule-exclusion create/update/delete, and any asset-management group create/delete/move/rename endpoints. These are mutation/provisioning actions out of scope for a read-only monitoring connector and can be added as a follow-up.
The incident-management API also exposes a separate /v2/incidents case-management surface (list/get/close/reopen, comments) alongside detections and detection-groups; it is not covered by this v1 surface either, for the same reason.
Development
npm install
npm run build
npm test
npm run lint # tsc --noEmitDocker
docker build -t eset-protect-mcp .
docker run -p 8080:8080 \
-e ESETPROTECT_CLIENT_ID=... \
-e ESETPROTECT_CLIENT_SECRET=... \
eset-protect-mcpAvailable Tools
26 toolsesetprotect_batch_get_detectionsA
Get multiple v2 detections by UUID in one call (max recommended 100, hard limit 1000).
Inputs: detectionUuids (required).
esetprotect_batch_get_devicesA
Get multiple managed devices by UUID in one call (max recommended 100, hard limit 1000).
Inputs: devicesUuids (required).
esetprotect_batch_get_usersA
Get multiple users by UUID in one call. Atomic: every requested user is returned or none.
Inputs: usersUuids (required).
esetprotect_get_detectionA
Get a single detection by UUID (v1 - legacy surface).
Inputs: detectionUuid (required).
esetprotect_get_detection_groupB
Get a single detection group by UUID.
Inputs: detectionGroupUuid (required).
esetprotect_get_detection_v2C
Get a single detection by UUID (v2).
Inputs: detectionUuid (required).
esetprotect_get_deviceA
Get a single managed device by UUID.
Inputs: deviceUuid (required).
esetprotect_get_edr_ruleA
Get a single EDR rule by UUID.
Inputs: ruleUuid (required).
esetprotect_get_edr_rule_exclusionA
Get a single EDR rule exclusion by UUID.
Inputs: exclusionUuid (required).
esetprotect_get_userC
Get a single user by UUID.
Inputs: userUuid (required).
esetprotect_list_detection_groupsC
List detection groups - related detections clustered together (e.g. the same malware across many devices).
Inputs: endTime, pageSize, pageToken, startTime, +2 more.
esetprotect_list_detectionsB
List detections (v1 - legacy surface, endpoint antivirus/firewall detections only). Prefer esetprotect_list_detections_v2 for new integrations.
Inputs: endTime, pageSize, pageToken, startTime, +1 more.
esetprotect_list_detections_v2B
List detections (v2 - covers endpoint, cloud office, and EDR detections).
Inputs: endTime, pageSize, pageToken, startTime, +1 more.
esetprotect_list_device_groupsB
List ESET PROTECT device (static/dynamic) groups.
Inputs: pageSize, pageToken.
esetprotect_list_device_os_vulnerabilitiesC
List operating-system vulnerabilities found on devices.
Inputs: pageSize, pageToken, deviceUuid, deviceGroupUuid.
esetprotect_list_device_patchesB
List patches (available/applied) for devices, optionally filtered by device, group, or patch type.
Inputs: pageSize, pageToken, patchType, deviceUuid, +1 more.
esetprotect_list_devicesB
List managed devices (endpoints), optionally filtered by display name, functionality status, or mute state.
Inputs: isMuted, pageSize, pageToken, displayNames, +1 more.
esetprotect_list_device_vulnerabilitiesC
List vulnerabilities found on devices (application, OS, or package scope).
Inputs: pageSize, pageToken, deviceUuid, deviceGroupUuid, +1 more.
esetprotect_list_edr_rule_exclusionsC
List EDR rule exclusions.
Inputs: pageSize, pageToken, includeTotalSize.
esetprotect_list_edr_rulesC
List EDR (Endpoint Detection & Response) detection rules.
Inputs: pageSize, pageToken, severityLevel, includeTotalSize.
esetprotect_list_group_devicesB
List the devices that are members of a given device group.
Inputs: groupUuid (required), pageSize, pageToken, recurseSubgroups.
esetprotect_list_patching_process_detailsC
List detailed patching-process history for devices within an optional time window.
Inputs: endTime, pageSize, pageToken, startTime, +2 more.
esetprotect_list_recent_application_patching_detailsB
List details of the most recent application-patching processes across the fleet.
Inputs: no named parameters.
esetprotect_list_recent_scansC
List details of recent vulnerability scans.
Inputs: pageSize, pageToken, deviceUuid, deviceGroupUuid.
esetprotect_list_usersA
List ESET Business Account users provisioned into this ESET PROTECT tenant, with optional filters.
Inputs: email, pageSize, pageToken, displayName, +4 more.
esetprotect_list_vulnerable_devicesC
List devices that currently have one or more known vulnerabilities.
Inputs: pageSize, pageToken, deviceGroupUuid.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 26 tools
The set follows a clear resource/action pattern, but several tools are easy to confuse: detections have v1 and v2 variants, device_os_vulnerabilities overlaps with device_vulnerabilities, and multiple patching/vulnerability-scan history tools exist. The descriptions help, but an agent must read carefully to avoid selecting the wrong list/get surface.
Every tool is prefixed with esetprotect_ and follows a consistent snake_case verb_noun pattern, including batch_get variants. There is no mixing of conventions or unexpected naming styles.
With 26 tools, the server is on the heavy side and above the ideal 3-15 range. Some consolidation is possible, especially around v1/v2 detections and multiple vulnerability/patching list tools, though the breadth is somewhat justified by the scope of ESET PROTECT.
The surface is almost entirely read-only: list/get/batch operations with no create, update, delete, or remediation tools. There are also no single-resource getters for device groups or vulnerability details, so some management and drill-down workflows dead-end.
Maintenance
Related MCP Connectors
Ask your AI assistant about your logs, traces, security events, robot runs and video. Read-only.
Read-only AI access to HAVN properties, leads, tasks, files, media, and analytics.
- DNS-DATA RemoteDeprecatedcom.dns-data
Connect AI assistants to authorized Windows and macOS devices for files, commands, and management.
Read-only AI access to Osboon business card analytics, viewers, links, connections and contacts.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to interact with Mender IoT platform for device management, deployment monitoring, and fleet analysis through natural language commands. Provides read-only access to device status, deployment logs, releases, and system monitoring capabilities.2MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to query an organization's device posture, compliance, policies, and software/CVE status through a remote, read-only, OAuth-secured MCP endpoint.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to read protected computers and their check-ins, computer groups, approval requests, the unified audit log, and child organizations through a zero-trust endpoint security portal API. Every operation is read-only, with credentials supplied per request as headers and never stored.Apache 2.0
- AlicenseAqualityFmaintenanceEnables an AI assistant to perform read-only defensive security lookups and analysis, including privacy-preserving breach checks, CVE and MITRE ATT&CK queries, log parsing, IOC defanging/refanging, hashing, and Windows event ID explanations.9MIT