Cortex XSIAM MCP for AI. Automate Incident Containment & Threat Response
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Cortex XSIAM connects your AI agent to a full Security Operations Center (SOC) platform. It lets you investigate incidents, contain threats on endpoints, and automate complex response playbooks—all from conversation.
Stop manually jumping between dashboards; get real-time alerts, indicators of compromise, and endpoint status right where you're working.
What your AI can do
Execute playbook
Runs an automated, defined incident response workflow using a playbook name and optional inputs.
Get alerts
Lists all security alerts detected by Cortex XSIAM so you can review recent threat activity.
Get endpoints
Retrieves a list of managed hosts and devices in the environment for asset auditing or targeting.
List all current security incidents, get full details on specific cases, or review recent detection rules firing.
Check the status of managed hosts, scan for malware, and immediately isolate a compromised device from the network.
Search raw logs across endpoints and networks using advanced queries, or list known indicators of compromise (IOCs).
Execute predefined incident response playbooks to handle tasks like blocking IPs or resetting passwords automatically.
Ask an AI about this
Cortex XSIAM MCP: 9 Tools for Security Operations
These nine tools allow your AI agent to manage the entire security lifecycle, from listing initial alerts to executing complex containment and remediation playbooks.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Cortex XSIAM on VinkiusExecute Playbook
Runs an automated, defined incident response workflow using a playbook name and optional inputs.
Get Alerts
Lists all security alerts detected by Cortex XSIAM so you can review recent threat...
Get Endpoints
Retrieves a list of managed hosts and devices in the environment for asset auditing...
Get Incident Details
Pulls deep, specific information about a single security incident using its ID.
Get Incidents
Lists all active or historical security incidents in Cortex XSIAM for workload...
Get Indicators
Checks the threat intelligence database to list known indicators of compromise (IOCs) related to a threat.
Isolate Endpoint
Immediately disconnects a compromised endpoint from the network using its unique ID.
Run Xql Query
Executes custom queries across logs and network data to perform advanced, targeted...
Scan Endpoint
Triggers a malware scan (quick or deep) on a specific endpoint ID to verify its...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Cortex XSIAM integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Cortex XSIAM, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Cortex XSIAM. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 9 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
The Messy Process of Incident Triage Today
Right. So, when a high-severity alert pops up, what happens? You open the dashboard, grab an IP address from one panel, copy it into another tool to check threat feeds, then you jump over to the asset inventory console just to find the affected host ID. It’s a cycle of clicking, copying, and cross-referencing dozens of separate tabs.
With this MCP, that manual handoff disappears. You tell your agent what happened in one prompt. It pulls together the incident context, checks the indicators, finds the impacted endpoints, and presents you with a single, unified view—no copy-pasting required.
Automating Containment with Cortex XSIAM
The manual steps that vanish are: 1) Finding the endpoint ID from a list of alerts. 2) Logging into the isolation console. 3) Manually executing the quarantine command. All those friction points disappear.
You don't just get data; you execute policy. You confirm the threat, and with one prompt, your agent performs the containment action using `isolate_endpoint`. It’s immediate, auditable, and drastically faster.
What your AI can actually do with this
Dealing with a high-severity alert is a race against time. Instead of opening five different consoles to gather context, this MCP lets your AI agent handle the initial investigation. You can ask it to list all related security alerts or check if specific indicators are known threats. The system then gathers endpoint data and incident details automatically, feeding you only what matters.
If you confirm a threat, you don't stop at reading; you tell the agent to run an automated response playbook. It handles everything from enriching compromised IPs to isolating endpoints—all orchestrated through your AI client. This capability means deep visibility into network activity and endpoint health without needing a security expert on standby.
019d757c-f778-72e7-9b7d-89d68bc4c236 
Here's how it actually works
The bottom line is that your AI client acts as the central nervous system, coordinating checks across multiple security tools so you don't have to.
Start by asking your agent to list recent security incidents or alerts using the available tools.
Tell the agent which endpoints or indicators need checking. It will gather current asset status and context for you.
Finally, instruct the agent to run a specific action, like isolating an endpoint or executing a predefined response playbook.
Who is this actually for?
This is for the SOC Analyst who gets tired of dashboard fatigue at 2 AM. It's for the Incident Responder who needs immediate containment options, and the Security Engineer building automated playbooks that need real-world data feeds.
They use this MCP to triage high-priority alerts by running get_alerts followed by run_xql_query to confirm the scope and source of a threat.
Needs quick, actionable commands. They use this MCP to gather context with get_incident_details and execute containment measures like isolate_endpoint immediately.
Builds automated workflows using execute_playbook, ensuring that complex, multi-step responses happen consistently every time an alert fires.
What Changes When You Connect
Instead of manually running a query, you ask your agent to run_xql_query and receive filtered results for threat hunting. This saves hours of manual data sifting.
You move past simple alerts. By using get_incident_details, the system provides all necessary context—who, what, and when—before you commit to an action.
Need to stop lateral movement fast? You can isolate a machine instantly with isolate_endpoint based on AI analysis of initial findings.
When dealing with repetitive tasks, use execute_playbook. It runs the entire response (enriching IOCs, blocking IPs) without you lifting a finger after setting it up once.
You get a clear picture of your assets by listing endpoints using get_endpoints, which is essential before running any remediation actions like scan_endpoint.
See it in action
A suspicious IP address is seen in an alert.
The analyst prompts the agent: 'Check this IP and see if it's a known bad actor.' The agent first uses get_indicators to check threat feeds, then uses run_xql_query against network logs for that specific IP. The result is immediate confirmation or denial.
A single user account shows unusual activity.
The agent gathers the full picture by calling get_incidents to list related cases, then uses get_endpoints to see which machines that user has logged into recently. This builds a scope of compromise much faster than clicking through dashboards.
A machine is suspected of being infected.
The analyst first gets the endpoint ID using get_endpoints. Then, they ask the agent to run a deep scan via scan_endpoint and immediately follow up with isolate_endpoint if the results are positive.
A known threat pattern emerges.
The team needs to replicate a response. Instead of manually running multiple steps, they trigger the predefined 'Malware Containment' playbook using execute_playbook, letting the system handle all the technical cleanup.
The honest tradeoffs
Over-relying on surface data
The analyst just reviews a list of alerts from get_alerts and assumes every alert requires manual investigation, wasting hours checking false positives.
First, use get_incidents to scope the problem. Then, instruct your agent to run run_xql_query against those limited results. This focuses deep investigation only on high-fidelity, correlated data.
Ignoring endpoint context
Running a cleanup playbook without first verifying the affected assets by running get_endpoints. You might accidentally target or miss critical machines.
Always start by listing your managed endpoints using get_endpoints. Use this list to confirm that all necessary hosts are in scope before triggering any containment action.
Jumping straight to isolation
The instinct is to hit 'Isolate' as soon as an alert fires. This risks disrupting critical business processes because you haven't confirmed the threat yet.
Before running isolate_endpoint, confirm the severity and indicators first. Use get_indicators and get_incident_details to build a case, making sure isolation is a confirmed last step.
When It Fits, When It Doesn't
Use this MCP if your core problem is orchestrating complex actions across multiple security tools—it's built for the Incident Response playbook. You need to correlate an alert with endpoint data and then run an automated fix. Don't use it if you just need to search a single log source; in that case, a dedicated query tool might be better. Also, don't assume this handles identity management alone; while it can reset passwords via playbooks, for pure user lifecycle management, look at specialized Identity Access Management tools instead. This MCP shines when the investigation requires moving from context gathering (get_incidents, get_alerts) to targeted action (isolate_endpoint, execute_playbook).
Questions you might have
How do I find out what happened after an alert using get_alerts? +
You first use get_alerts to see which rules fired. Then, you must run run_xql_query immediately afterward, filtering the query by the specific alert ID you found. This gives you the raw data behind the warning.
Can I automate a full response using execute_playbook? +
Yes. execute_playbook runs complex workflows automatically. You just need to provide the playbook name and any required inputs, letting the MCP handle the multi-step actions.
What do I use if an endpoint is infected? Should I use scan_endpoint or isolate_endpoint? +
It's a two-step process. First, run scan_endpoint to confirm the infection and get proof of life. Only after confirming the threat should you then call isolate_endpoint for containment.
How can I check if an indicator is suspicious using get_indicators? +
You use get_indicators by providing a hash, IP, or domain. This tool checks known threat intelligence sources and tells you if the artifact has been marked as malicious.
What's the proper way to structure a deep investigation query using `run_xql_query`? +
You need to specify your data sources and necessary filters within the XQL string. The tool returns structured results from logs, network traffic, and endpoint data, letting you correlate multiple events in one go.
I'm setting up a new environment; how do I use `get_endpoints` to check my full device inventory? +
Running get_endpoints lists every managed host and device linked to your Cortex XSIAM. This helps you quickly audit coverage, ensuring no machines are offline or unmonitored.
If I only have a general idea of a threat, how can I use `get_incident_details` for context? +
You must provide the specific incident ID to get details. This tool pulls deep information about that single event, giving you the full background needed before deciding on next steps.
When tracking a high volume of alerts, how do I use `get_incidents` to focus only on critical cases? +
You can use parameters with get_incidents to sort and limit results. This lets you narrow down the list to specific severity levels or timeframes, managing analyst workload efficiently.
We've already built the connector for Cortex XSIAM. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 9 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.