Connectors · MSP & PSA
Microsoft Sentinel MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Microsoft Sentinel and the rest of your systems. It gives your agent 98 Microsoft Sentinel actions (54 reads, 23 writes and 21 that wait for a confirmation) and holds the Microsoft Sentinel credential itself, so the agent never does. It paces Microsoft Sentinel's limits and records every call.
100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.
claude mcp add --transport http actionlake https://connect.actionlake.com/mcpdo { action: "azure_sentinel.operations.list", intent: "check which Microsoft Sentinel account is connected"}# Lists all operations available Azure Security Insights Resource Provider · read · 1 upstream call- 98actionsin one signed file
- 54readlook, never change
- 23writecreate and update
- 21wait for a yesdestructive, confirmed first
Set up
Connect Microsoft Sentinel in three steps.
- 01
Add actionlake to Claude.
Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
https://connect.actionlake.com/mcp - 02
Connect Microsoft Sentinel.
OAuth 2.0, client credentials. Paste the app's client id and secret; actionlake fetches and renews the token, and nobody has to sign in. actionlake stores what you paste write-only, sealed under your tenant's key.
client_idEntra application (client) id of a service principal with a Microsoft Sentinel role on the workspace resource groupclient_secretEntra client secrettenant_idYour Microsoft Sentinel tenant id
account { op: "connections.requirements", service: "azure_sentinel" } # free: exactly what to paste, and what is found for you - 03
Make the first call. Check before a write.
Ask in plain words; Claude picks the action. Before anything changes,
checkvalidates the call and prices it, for free.do { action: "azure_sentinel.operations.list" }# lists all operations available azure security insights resource provider · read · 1 upstream callcheck { action: "azure_sentinel.alert_rules.delete", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Microsoft Sentinel at a glance
- Sign-in
- OAuth 2.0, client credentials. Paste the app's client id and secret; actionlake fetches and renews the token, and nobody has to sign in.
- You provide
client_idclient_secret- Setup fields
tenant_id- Limits we pace to
- 200 calls a minute, per connection
- API covered
- Microsoft Sentinel (incidents, comments, tasks, analytics rules, automation rules, watchlists, threat intelligence, data connectors, content hub) API Azure Resource Manager Microsoft.SecurityInsights stable/2025-09-01/openapi.json (2025-09-01)
management.azure.com - Built from
- Microsoft Sentinel's documentation, read 23 October 2026OpenAPI spec (GitHub)learn.microsoft.com
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 49 of 54 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 1, so ask can line Microsoft Sentinel up with your other systems
- Data
- Processed and stored in Sydney, Australia. No US region yet
Capabilities
What an AI agent can do in Microsoft Sentinel.
98 actions in 27 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 54readlook, never change
- 23writecreate and update
- 21destructivewait for a confirmation
Threat intelligence indicators
94 read · 4 write · 1 destructive
Get all threat intelligence indicators · View a threat intelligence indicator by name · Get threat intelligence indicators metrics (Indicators counts by Type, Threat Type, Source) · Create a new threat intelligence indicator
Incidents
85 read · 1 write · 2 destructive
Gets all incidents · Gets a given incident · Deletes a given incident · Gets all alerts for an incident
Metadata
52 read · 2 write · 1 destructive
List of all metadata · Get a Metadata · Update an existing Metadata · Delete a Metadata
Source controls
53 read · 1 write · 1 destructive
Gets all source controls, without source control items · Gets a list of repositories metadata · Gets a source control byt its identifier · Delete a source control
Alert rule actions
42 read · 1 write · 1 destructive
Gets all actions of alert rule · Gets the action of alert rule · Delete the action of alert rule · Creates or updates the action of alert rule
Alert rules
42 read · 1 write · 1 destructive
Gets all alert rules · Gets the alert rule · Delete the alert rule · Creates or updates the alert rule
Automation rules
42 read · 1 write · 1 destructive
Gets all automation rules · Gets the automation rule · Delete the automation rule · Creates or updates the automation rule
Bookmarks
42 read · 1 write · 1 destructive
Gets all bookmarks · Gets a bookmark · Delete the bookmark · Creates or updates the bookmark
19 more areas · 55 actions
- Content packages 4
- Content templates 4
- Data connector definitions 4
- Data connectors 4
- Incident comments 4
- Incident relations 4
- Incident tasks 4
- Onboarding states 4
- Security ML analytics settings 4
- Watchlist items 4
- Watchlists 4
- Alert rule templates 2
- Product packages 2
- Product templates 2
- Entities 1
- Operations 1
- Resource groups 1
- Subscriptions 1
- Workspaces 1
Top actions
The Microsoft Sentinel API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 18 of Microsoft Sentinel's 19 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
azure_sentinel.incidents.listGets all incidents | Gets all incidents. | read | security.incident.list17 connectors |
azure_sentinel.alert_rules.listGets all alert rules | Gets all alert rules. | read | security.detection_rule.list13 connectors |
azure_sentinel.incident_comments.listGets all comments for a given incident | Gets all comments for a given incident. | read | security.incident_note.list9 connectors |
azure_sentinel.threat_intelligence_indicators.listGet all threat intelligence indicators | Get all threat intelligence indicators. | read | security.threat.list7 connectors |
azure_sentinel.incidents.getGets a given incident | Gets a given incident. | read | security.incident.get17 connectors |
azure_sentinel.automation_rules.listGets all automation rules | Gets all automation rules. | read | automation.rule.list22 connectors |
azure_sentinel.alert_rules.getGets the alert rule | Gets the alert rule. | read | security.detection_rule.get11 connectors |
azure_sentinel.incident_comments.getGets an incident comment | Gets an incident comment. | read | security.incident_note.get6 connectors |
azure_sentinel.alert_rules.deleteDelete the alert rule | Delete the alert rule. | destructive | security.detection_rule.delete11 connectors |
azure_sentinel.incidents.create_or_updateCreates or updates an incident | Creates or updates an incident. | write | security.incident.upsert2 connectors |
azure_sentinel.alert_rules.create_or_updateCreates or updates the alert rule | Creates or updates the alert rule. | write | security.detection_rule.upsert2 connectors |
azure_sentinel.incident_comments.create_or_updateCreates or updates a comment for a given incident | Creates or updates a comment for a given incident. | write | this connector |
Try it
Prompts to paste into Claude.
Each one names the Microsoft Sentinel actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Which Microsoft Sentinel incidents are open, and how long has each been open?
calls
azure_sentinel.incidents.list - Writewrite
Creates or updates an incident in Microsoft Sentinel. Show me exactly what you will send before you save it.
calls
azure_sentinel.incidents.create_or_update
Governance for Microsoft Sentinel
Destructive actions wait for a yes.
21 Microsoft Sentinel actions are destructive. These are all of them:
- Delete the alert rule
azure_sentinel.alert_rules.delete - Deletes a comment for a given incident
azure_sentinel.incident_comments.delete - Deletes a given incident
azure_sentinel.incidents.delete - Delete the automation rule
azure_sentinel.automation_rules.delete - Triggers playbook on a specific entity
azure_sentinel.entities.run_playbook - Triggers playbook on a specific incident
azure_sentinel.incidents.run_playbook - Delete a Metadata
azure_sentinel.metadata.delete - Delete the bookmark
azure_sentinel.bookmarks.delete - Delete a watchlist
azure_sentinel.watchlists.delete - Delete the incident task
azure_sentinel.incident_tasks.delete - Delete the data connector
azure_sentinel.data_connectors.delete - Delete a source control
azure_sentinel.source_controls.delete - Delete a watchlist item
azure_sentinel.watchlist_items.delete - Delete an installed template
azure_sentinel.content_templates.delete - Delete Sentinel onboarding state
azure_sentinel.onboarding_states.delete - Delete the action of alert rule
azure_sentinel.alert_rule_actions.delete - Deletes a relation for a given incident
azure_sentinel.incident_relations.delete - Uninstall a package from the workspace
azure_sentinel.content_packages.uninstall - Delete the data connector definition
azure_sentinel.data_connector_definitions.delete - Delete the Security ML Analytics Settings
azure_sentinel.security_ml_analytics_settings.delete - Delete a threat intelligence indicator
azure_sentinel.threat_intelligence_indicators.delete
A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.
Delete the alert rule?
intent: the user asked me to delete the alert rule
account { op: "keys.create", name: "azure-sentinel-agent", allowed_actions: ["azure_sentinel.*"], cap_calls: 2000, confirm_policy: "human_for_writes", approval_rules: [ { match: "azure_sentinel.alert_rules.delete", approver: "[email protected]" }, { match: "azure_sentinel.incident_comments.delete", approver: "[email protected]" }, { match: "azure_sentinel.incidents.delete", approver: "[email protected]" } ]}Cost
What Microsoft Sentinel costs on actionlake.
Every Microsoft Sentinel action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Microsoft Sentinel refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
- 1upstream call per actionall 98 Microsoft Sentinel actions
- 24actions pageeach page of a long list is one call
- 49reads can be cacheda cache hit costs nothing
- 100,000calls a month, freepaid plans from US$49 a month for 250,000
Action index
Every Microsoft Sentinel action.
98 actions by area: the id your agent calls, what it does, and its effect.
Threat intelligence indicators9
azure_sentinel.threat_intelligence_indicators.append_tagsAppend tags to a threat intelligence indicatorwriteazure_sentinel.threat_intelligence_indicators.createUpdate a threat Intelligence indicatorwriteazure_sentinel.threat_intelligence_indicators.create_indicatorCreate a new threat intelligence indicatorwriteazure_sentinel.threat_intelligence_indicators.deleteDelete a threat intelligence indicatordestructiveazure_sentinel.threat_intelligence_indicators.getView a threat intelligence indicator by namereadazure_sentinel.threat_intelligence_indicators.listGet all threat intelligence indicatorsreadazure_sentinel.threat_intelligence_indicators.list_metricsGet threat intelligence indicators metrics (Indicators counts by Type, Threat Type, Source)readazure_sentinel.threat_intelligence_indicators.query_indicatorsQuery threat intelligence indicators as per filtering criteriareadazure_sentinel.threat_intelligence_indicators.replace_tagsReplace tags added to a threat intelligence indicatorwrite
Incidents8
azure_sentinel.incidents.create_or_updateCreates or updates an incidentwriteazure_sentinel.incidents.deleteDeletes a given incidentdestructiveazure_sentinel.incidents.getGets a given incidentreadazure_sentinel.incidents.listGets all incidentsreadazure_sentinel.incidents.list_alertsGets all alerts for an incidentreadazure_sentinel.incidents.list_bookmarksGets all bookmarks for an incidentreadazure_sentinel.incidents.list_entitiesGets all entities for an incidentreadazure_sentinel.incidents.run_playbookTriggers playbook on a specific incidentdestructive
Metadata5
azure_sentinel.metadata.createCreate a Metadatawriteazure_sentinel.metadata.deleteDelete a Metadatadestructiveazure_sentinel.metadata.getGet a Metadatareadazure_sentinel.metadata.listList of all metadatareadazure_sentinel.metadata.updateUpdate an existing Metadatawrite
Source controls5
azure_sentinel.source_controls.createCreates a source controlwriteazure_sentinel.source_controls.deleteDelete a source controldestructiveazure_sentinel.source_controls.getGets a source control byt its identifierreadazure_sentinel.source_controls.listGets all source controls, without source control itemsreadazure_sentinel.source_controls.list_repositoriesGets a list of repositories metadataread
Alert rule actions4
azure_sentinel.alert_rule_actions.create_or_updateCreates or updates the action of alert rulewriteazure_sentinel.alert_rule_actions.deleteDelete the action of alert ruledestructiveazure_sentinel.alert_rule_actions.getGets the action of alert rulereadazure_sentinel.alert_rule_actions.listGets all actions of alert ruleread
Alert rules4
azure_sentinel.alert_rules.create_or_updateCreates or updates the alert rulewriteazure_sentinel.alert_rules.deleteDelete the alert ruledestructiveazure_sentinel.alert_rules.getGets the alert rulereadazure_sentinel.alert_rules.listGets all alert rulesread
Automation rules4
azure_sentinel.automation_rules.create_or_updateCreates or updates the automation rulewriteazure_sentinel.automation_rules.deleteDelete the automation ruledestructiveazure_sentinel.automation_rules.getGets the automation rulereadazure_sentinel.automation_rules.listGets all automation rulesread
Bookmarks4
azure_sentinel.bookmarks.create_or_updateCreates or updates the bookmarkwriteazure_sentinel.bookmarks.deleteDelete the bookmarkdestructiveazure_sentinel.bookmarks.getGets a bookmarkreadazure_sentinel.bookmarks.listGets all bookmarksread
Content packages4
azure_sentinel.content_packages.getGets an installed packages by its idreadazure_sentinel.content_packages.installInstall a package to the workspacewriteazure_sentinel.content_packages.listGets all installed packagesreadazure_sentinel.content_packages.uninstallUninstall a package from the workspacedestructive
Content templates4
azure_sentinel.content_templates.deleteDelete an installed templatedestructiveazure_sentinel.content_templates.getGets a template byt its identifierreadazure_sentinel.content_templates.installInstall a templatewriteazure_sentinel.content_templates.listGets all installed templatesread
Data connector definitions4
azure_sentinel.data_connector_definitions.create_or_updateCreates or updates the data connector definitionwriteazure_sentinel.data_connector_definitions.deleteDelete the data connector definitiondestructiveazure_sentinel.data_connector_definitions.getGets a data connector definitionreadazure_sentinel.data_connector_definitions.listGets all data connector definitionsread
Data connectors4
azure_sentinel.data_connectors.create_or_updateCreates or updates the data connectorwriteazure_sentinel.data_connectors.deleteDelete the data connectordestructiveazure_sentinel.data_connectors.getGets a data connectorreadazure_sentinel.data_connectors.listGets all data connectorsread
Incident comments4
azure_sentinel.incident_comments.create_or_updateCreates or updates a comment for a given incidentwriteazure_sentinel.incident_comments.deleteDeletes a comment for a given incidentdestructiveazure_sentinel.incident_comments.getGets an incident commentreadazure_sentinel.incident_comments.listGets all comments for a given incidentread
Incident relations4
azure_sentinel.incident_relations.create_or_updateCreates or updates the incident relationwriteazure_sentinel.incident_relations.deleteDeletes a relation for a given incidentdestructiveazure_sentinel.incident_relations.getGets a relation for a given incidentreadazure_sentinel.incident_relations.listGets all relations for a given incidentread
Incident tasks4
azure_sentinel.incident_tasks.create_or_updateCreates or updates the incident taskwriteazure_sentinel.incident_tasks.deleteDelete the incident taskdestructiveazure_sentinel.incident_tasks.getGets an incident taskreadazure_sentinel.incident_tasks.listGets all incident tasksread
Onboarding states4
azure_sentinel.onboarding_states.createCreate Sentinel onboarding statewriteazure_sentinel.onboarding_states.deleteDelete Sentinel onboarding statedestructiveazure_sentinel.onboarding_states.getGet Sentinel onboarding statereadazure_sentinel.onboarding_states.listGets all Sentinel onboarding statesread
Security ML analytics settings4
azure_sentinel.security_ml_analytics_settings.create_or_updateCreates or updates the Security ML Analytics Settingswriteazure_sentinel.security_ml_analytics_settings.deleteDelete the Security ML Analytics Settingsdestructiveazure_sentinel.security_ml_analytics_settings.getGets the Security ML Analytics Settingsreadazure_sentinel.security_ml_analytics_settings.listGets all Security ML Analytics Settingsread
Watchlist items4
azure_sentinel.watchlist_items.create_or_updateCreate or update a watchlist itemwriteazure_sentinel.watchlist_items.deleteDelete a watchlist itemdestructiveazure_sentinel.watchlist_items.getGet a watchlist itemreadazure_sentinel.watchlist_items.listGet all watchlist Itemsread
Watchlists4
azure_sentinel.watchlists.create_or_updateCreate or update a Watchlist and its Watchlist Items (bulk creation, e.gwriteazure_sentinel.watchlists.deleteDelete a watchlistdestructiveazure_sentinel.watchlists.getGet a watchlist, without its watchlist itemsreadazure_sentinel.watchlists.listGet all watchlists, without watchlist itemsread
Alert rule templates2
azure_sentinel.alert_rule_templates.getGets the alert rule templatereadazure_sentinel.alert_rule_templates.listGets all alert rule templatesread
Product packages2
azure_sentinel.product_packages.getGets a package by its identifier from the catalogreadazure_sentinel.product_packages.listGets all packages from the catalogread
Product templates2
azure_sentinel.product_templates.getGets a template by its identifierreadazure_sentinel.product_templates.listGets all templates in the catalogread
Entities1
azure_sentinel.entities.run_playbookTriggers playbook on a specific entitydestructive
Operations1
azure_sentinel.operations.listLists all operations available Azure Security Insights Resource Providerread
Resource groups1
azure_sentinel.resource_groups.listList the resource groups in a subscriptionread
Subscriptions1
azure_sentinel.subscriptions.listList the subscriptions the credential can seeread
Workspaces1
azure_sentinel.workspaces.listList the Log Analytics workspaces in a subscriptionread
Questions
The Microsoft Sentinel MCP server, answered.
Is there a Microsoft Sentinel MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 98 Microsoft Sentinel actions across threat intelligence indicators, incidents, metadata and source controls, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Microsoft Sentinel.
What do I need to connect Microsoft Sentinel?
Entra application (client) id of a service principal with a Microsoft Sentinel role on the workspace resource group and Entra client secret. Plus your tenant id. The free call account { op: "connections.requirements", service: "azure_sentinel" } lists exactly what to paste.
Can an AI agent delete or change things in Microsoft Sentinel?
Only if its key allows it, and only with an explicit confirmation. 21 Microsoft Sentinel actions are destructive, including delete the alert rule, deletes a comment for a given incident and deletes a given incident. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.
Will an AI agent use up my Microsoft Sentinel API limit?
actionlake paces every call to the limits in the Microsoft Sentinel service file: 200 calls a minute per connection. Repeated reads are answered from cache (49 Microsoft Sentinel actions can be), which never reaches Microsoft Sentinel.
What does a Microsoft Sentinel action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (24 Microsoft Sentinel actions page). Cache hits, searches, dry runs and calls Microsoft Sentinel refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
Which Microsoft Sentinel API does it cover?
The Microsoft Sentinel (incidents, comments, tasks, analytics rules, automation rules, watchlists, threat intelligence, data connectors, content hub) API Azure Resource Manager Microsoft.SecurityInsights stable/2025-09-01/openapi.json (2025-09-01) at management.azure.com. It is built from Microsoft Sentinel's documentation as read on 23 October 2026. 98 actions in 27 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Microsoft Sentinel data, and where?
actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Microsoft Sentinel credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.
Which AI clients can use it?
Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.
Give your agent Microsoft Sentinel today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.
