SentinelOne MCP Server
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., "@SentinelOne MCP Server Summarize threats by site for the past week"
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.
SentinelOne MCP Server
Read-only MCP server for the SentinelOne management API (v2.1). Exposes sites, endpoint agents, threats and the activity log to Claude (Claude Desktop / Cowork, Claude Code, or any MCP client), with aggregation tools built for MSP/MSSP reporting across a site-segmented client base.
Tools
Tool | Purpose |
| Sites (how clients are segmented in an MSP console) with licence counts, SKU, expiry and health |
| Estate-wide endpoint health by site: agents, online/offline, infected, needs-update, reboot-required, stale, OS mix — the health reporting tool |
| Endpoint drill-down: per-machine status, version, last active, last user (filterable by site, health flags, OS) |
| Threat drill-down: name, classification, verdict, incident/mitigation status, affected machine (filterable by site, status, date range) |
| Aggregated threat counts by site, verdict, classification, top threats and machines — the periodic reporting tool |
| Console activity/audit log: mitigations, installs, policy changes, logins |
| Raw GET to any other SentinelOne endpoint — disabled unless |
The tool set is deliberately lean, focused on endpoint health and threat reporting across clients. The raw API tool covers anything else (groups, exclusions, installed applications, STAR alerts, system info) but is off by default so shared deployments stay strictly read-only.
Related MCP server: redash-mcp
Quick start (Claude Desktop / Cowork)
Add to mcpServers in your claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json):
"sentinelone": {
"command": "npx",
"args": ["-y", "sentinelone-mcp"],
"env": {
"S1_API_TOKEN": "your-api-token",
"S1_BASE_URL": "https://your-console.sentinelone.net"
}
}For Claude Code:
claude mcp add-json sentinelone --scope user '{"type":"stdio","command":"npx","args":["-y","sentinelone-mcp"],"env":{"S1_API_TOKEN":"your-api-token","S1_BASE_URL":"https://your-console.sentinelone.net"}}'Requires Node.js 18+.
Configuration
Variable | Value |
| API token — generate in the console under My User → Actions → API Token Operations |
| Your management console URL, e.g. |
| Optional — set to |
Notes on tokens:
API tokens expire on a schedule set in the console (commonly 30 days, extendable under Settings → Users, or use a service user for longer-lived tokens). A 401 from this server almost always means the token has expired — regenerate it.
The token inherits the scope and role of the user who generated it. A Viewer-role user scoped to the account is ideal: the server only ever issues GET requests, but a least-privilege token means mistakes elsewhere can't mutate anything.
Each team member should generate their own token so console audit logs attribute API reads correctly.
Reporting across a client base
Sites are how clients are segmented in an MSP/MSSP console. The intended flow for "state of our clients' endpoints" reporting:
s1_list_sites— the client list, with licence usage and expiry per site.s1_estate_overview— one call sweeps every agent and returns a site-by-site health table: agent count, online/offline, infected, agents needing an update, reboot-required, stale agents (default: not seen in 30 days) and OS mix. Aggregation happens inside the server, so thousands of agents come back as a few KB of counts.s1_threat_summarywithcreated_after— threats for the reporting period, grouped by site with verdict/classification breakdowns, top threat names and noisiest machines.s1_agents/s1_threats— drill into the individual machines or detections behind any number, thens1_activitiesfor the audit trail of what was done about them.
Development
git clone https://github.com/jamescash1976/sentinelone-mcp.git
cd sentinelone-mcp
npm install
cp .env.example .env # fill in token + console URL
npm run dev # watch mode
npm run inspector # MCP Inspector UI
npm run build # compile to dist/Notes
Strictly read-only: every tool issues GET requests only. Mitigation, policy and agent actions are deliberately not implemented.
Pagination is cursor-based throughout: responses include
nextCursor— pass it back ascursorto continue. The sweep tools (s1_estate_overview,s1_threat_summary) page automatically up to their safety caps and say so when truncated.Rate limits (HTTP 429) are retried automatically using the
Retry-Afterheader, up to 3 times.Compact projections drop the (many) null fields SentinelOne returns and exclude site registration tokens from output. Use
full_details: trueons1_agents/s1_threatswhen you need the complete raw records.Dates in filters are ISO 8601, e.g.
2026-07-01T00:00:00Z.
Available Tools
6 toolss1_activitiesA
Activity/audit log from the management console: threat mitigations, agent installs/upgrades, policy changes, user logins, exclusion changes and more, newest first. Useful as evidence of actions taken (e.g. "show what was done about the threat on machine X").
Inputs: limit, cursor, site_ids, agent_ids, +3 more.
s1_agentsA
List endpoint agents with filters — the drill-down tool for machines. Returns each agent's computer name, site, OS, agent version, online/infected/up-to-date status, last active date, last logged-in user and network status. Filter by site, name, health flags or OS. Returns a compact summary per agent by default; set full_details=true for complete raw records (large). Call s1_list_sites first to resolve site IDs.
Inputs: limit, cursor, infected, os_types, +10 more.
s1_estate_overviewA
Estate-wide endpoint health overview grouped by site — one call answers "what is the state of every client's endpoints?". Sweeps all agents server-side and returns a table per site: agent count, online/offline, infected, agents needing an update, reboot-required, stale agents (not seen for N days) and OS breakdown. This is the primary tool for MSP health/AV reporting. Use s1_agents to drill into the individual machines afterwards.
Inputs: site_ids, max_agents, stale_days.
s1_list_sitesA
List sites in the SentinelOne management console. In an MSP/MSSP console, sites are how clients are segmented — use this to discover the client list, then feed site IDs into the other tools. Returns name, state, licence counts (active/total), SKU/suite, expiration and health status per site.
Inputs: name, limit, cursor, states.
s1_threatsA
List threats/detections with filters — the drill-down tool for incidents. Returns each threat's name, classification, confidence level (malicious/suspicious), analyst verdict, incident status, mitigation status and actions, detection engine, file path/hash, affected machine and site, sorted newest first. Returns a compact summary per threat by default; set full_details=true for complete raw records. Use s1_threat_summary for aggregated counts across the estate.
Inputs: ids, limit, cursor, site_ids, +10 more.
s1_threat_summaryA
Aggregated threat counts grouped by site — the reporting/dashboard tool. Sweeps ALL matching threats server-side and returns compact counts: per site by incident status, analyst verdict and classification, plus estate-wide totals, the most common threat names and the machines with the most detections. Ideal for "threats this month across all clients" reporting — pass created_after for the period. Use s1_threats to drill into individual detections afterwards.
Inputs: site_ids, max_threats, created_after, created_before, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool has a distinct purpose: site discovery, estate health, agent detail, threat detail, threat aggregation, and activity logging. There is no overlap in functionality; even the overview/summary tools are clearly differentiated by granularity and use case.
All tools share the 's1_' prefix and snake_case, but the second part mixes verb+noun (s1_list_sites), noun+noun (s1_estate_overview, s1_threat_summary), and bare plural nouns (s1_agents, s1_threats, s1_activities). This is inconsistent but still readable and predictable enough to navigate.
With 6 tools, the server is well-scoped for its apparent purpose of read-only SentinelOne monitoring and reporting. Each tool covers a necessary facet of the domain without redundancy or bloat.
The read-only surface is fairly complete, covering sites, estate health, agents, threats, threat summaries, and activity logs. Minor gaps exist, such as lack of per-site detail beyond the list view and no direct single-threat lookup, but these can be worked around with existing filters and drill-downs.
Related MCP Connectors
Read-only MCP server for AIStatusDashboard status, incidents, metrics, and fallback recommendations.
Read-only MCP server for ClassQuill, a tutoring-business-management platform.
Free, read-only public-source threat intel over MCP. No account or API key.
Read-only MCP server for The Quiet Protocol's engines, benchmarks, proof, and business data.
Related MCP Servers
- AlicenseBqualityDmaintenanceMCP server for Signal Sciences (Fastly NGWAF) API, providing progressive disclosure and 40+ read-only tools for managing corporations and sites.481MIT
- FlicenseBqualityDmaintenanceRead-only MCP server wrapping the Redash REST API, enabling query listing, execution, data source inspection, and dashboard retrieval with optional per-user API key support.7-
- FlicenseNot gradedqualityCmaintenanceMCP server that exposes Acronis Cyber Protect Cloud APIs as 14 read-only tools for managing alerts, tasks, agents, resources, policies, and tenants.-
- AlicenseBqualityCmaintenanceA read-only MCP server for Microsoft Intune and Entra ID that enables list, get, search, and reporting operations for tenant visibility, audits, troubleshooting, and health reporting without write actions. It includes authentication helpers, report exports, and metadata discovery tools.361MIT