Connectors · Observability & analytics
Microsoft Fabric MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Microsoft Fabric and the rest of your systems. It gives your agent 762 Microsoft Fabric actions (336 reads, 307 writes and 119 that wait for a confirmation) and holds the Microsoft Fabric credential itself, so the agent never does. It paces Microsoft Fabric'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: "microsoft_fabric.workspaces.list_workspaces", intent: "check which Microsoft Fabric account is connected"}# Returns a list of workspaces the principal can access. Use the roles query parameter to filter results by the principal workspace role · read · 1 upstream call- 762actionsin one signed file
- 336readlook, never change
- 307writecreate and update
- 119wait for a yesdestructive, confirmed first
Set up
Connect Microsoft Fabric 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 Fabric.
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 an app registration allowed to call Fabric APIsclient_secretEntra client secrettenant_idYour Microsoft Fabric tenant id
account { op: "connections.requirements", service: "microsoft_fabric" } # 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: "microsoft_fabric.workspaces.list_workspaces" }# returns a list of workspaces the principal can access. use the roles query parameter to filter results by the principal workspace role · read · 1 upstream callcheck { action: "microsoft_fabric.report.create_report", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Microsoft Fabric 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
- 100 calls a minute, per connection
- API covered
- Fabric REST API v1 (fabric-rest-api-specs, Swagger 2.0)
api.fabric.microsoft.com/v1 - Built from
- Microsoft Fabric's documentation, read 22 October 2026GitHublearn.microsoft.com
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 278 of 336 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 4, so ask can line Microsoft Fabric 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 Fabric.
762 actions in 88 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 336readlook, never change
- 307writecreate and update
- 119destructivewait for a confirmation
Workspaces
3512 read · 15 write · 8 destructive
Returns a list of workspaces the principal can access. Use the roles query parameter to filter results by the principal workspace role · Returns a list of workspace role assignments · Returns specified workspace information · Creates a new workspace
Data agent
3115 read · 11 write · 5 destructive
Returns a list of DataAgents from the specified workspace · Returns schema elements at a level of the schema tree for a staging datasource · Returns a list of staging fewshots for a datasource of the specified DataAgent · Returns a list of staging datasources for the specified DataAgent
Apache airflow job
2411 read · 10 write · 3 destructive
List Apache Airflow pool templates · Returns a list of Apache Airflow job files from the specified Apache Airflow job · Returns a list of installed libraries for the specified Apache Airflow job environment · Returns a list of Apache Airflow jobs from the specified workspace
Environment
2311 read · 7 write · 5 destructive
Returns a list of environments from the specified workspace · Get a list of libraries staged into environment · Get environment published libraries · Returns properties of the specified environment
Lakehouse
239 read · 11 write · 3 destructive
Returns a list of lakehouses from the specified workspace · Returns a list of livy sessions from the specified item identifier · Returns a list of livy sessions from the specified item identifier (beta) · Returns a list of materialized lake views execution definitions for a lakehouse
Admin domains
196 read · 7 write · 6 destructive
Returns a list of the workspaces assigned to the specified domain · Returns a list of domain role assignments · Returns the specified domain info · Returns the specified domain info (preview)
Eventstream
188 read · 9 write · 1 destructive
Returns a list of eventstreams from the specified workspace · Returns properties of the specified eventstream · Returns the specified destination of the eventstream · Returns the connection information of a specified destination of the eventstream
Deployment pipelines
178 read · 5 write · 4 destructive
Returns a list of the up-to-20 most recent deploy operations performed on the specified deployment pipeline · Returns a list of deployment pipeline role assignments · Returns the supported items from the workspace assigned to the specified stage of the specified deployment pipeline · Returns the specified deployment pipeline stages
80 more areas · 572 actions
- Gateways 17
- Warehouse 17
- Items 16
- Policy set 16
- Spark 16
- Data pipeline 15
- Dataflow 15
- ML model 15
- Git 14
- SQL database 13
- Connections 12
- Mirrored catalog 12
- Notebook 12
- KQL database 11
- Mirrored azure databricks catalog 11
- Mirrored database 11
- Spark job definition 11
- Copy job 10
- Graph model 10
- Data build tool job 9
- Job scheduler 9
- Monitoring 9
- Admin tenants 8
- Admin workspaces 8
- Reflex 8
- Semantic model 8
- Anomaly detector 7
- Azure databricks storage 7
- Cosmos DB database 7
- Digital twin builder 7
- Digital twin builder flow 7
- Event schema set 7
- Eventhouse 7
- Graph query set 7
- Graphql API 7
- KQL dashboard 7
- KQL queryset 7
- Map 7
- Mounted data factory 7
- Ontology 7
- Operations agent 7
- Org app 7
- Org app audience 7
- Paginated report 7
- Plan 7
- Report 7
- Snowflake database 7
- User data function 7
- Variable library 7
- Folders 6
- One lake shortcuts 6
- SQL endpoint 6
- App backend 5
- External data shares provider 5
- ML experiment 5
- One lake data access security 5
- One lake settings 5
- Warehouse snapshot 5
- Admin tags 4
- Admin workloads 4
- Capacities 4
- Managed private endpoints 4
- Admin items 3
- Admin policy sets 3
- Recoverable items 3
- Tags 3
- Admin external data shares provider 2
- Admin labels 2
- Admin sharing links 2
- Domains 2
- External data shares recipient 2
- Long running operations 2
- One lake lifecycle policy 2
- Surge protection 2
- Admin users 1
- Catalog 1
- Dashboard 1
- Datamart 1
- Mirrored warehouse 1
- Real time intelligence 1
Top actions
The Microsoft Fabric API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 16 of Microsoft Fabric's 16 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
microsoft_fabric.report.list_reportsReturns a list of reports from the specified workspace | Returns a list of reports from the specified workspace. | read | analytics.report.list37 connectors |
microsoft_fabric.semantic_model.list_semantic_modelsReturns a list of semantic models from the specified workspace | Returns a list of semantic models from the specified workspace. | read | analytics.dataset.list28 connectors |
microsoft_fabric.workspaces.list_workspacesReturns a list of workspaces the principal can access. Use the roles query parameter to filter results by the principal workspace role | Returns a list of workspaces the principal can access. Use the roles query parameter to filter results by the principal workspace role. | read | work.workspace.list50 connectors |
microsoft_fabric.workspaces.list_workspace_role_assignmentsReturns a list of workspace role assignments | Returns a list of workspace role assignments. | read | work.workspace_member.list26 connectors |
microsoft_fabric.report.get_reportReturns properties of the specified report | Returns properties of the specified report. | read | analytics.report.get39 connectors |
microsoft_fabric.semantic_model.get_semantic_modelReturns properties of the specified semantic model | Returns properties of the specified semantic model. | read | analytics.dataset.get22 connectors |
microsoft_fabric.report.create_reportCreates a report in the specified workspace | Creates a report in the specified workspace. | write | analytics.report.create27 connectors |
microsoft_fabric.workspaces.get_workspaceReturns specified workspace information | Returns specified workspace information. | read | work.workspace.get41 connectors |
microsoft_fabric.semantic_model.create_semantic_modelCreates a semantic model in the specified workspace | Creates a semantic model in the specified workspace. | write | analytics.dataset.create21 connectors |
microsoft_fabric.workspaces.create_workspaceCreates a new workspace | Creates a new workspace. | write | work.workspace.create36 connectors |
microsoft_fabric.semantic_model.delete_semantic_modelDeletes the specified semantic model | Deletes the specified semantic model. | destructive | analytics.dataset.delete23 connectors |
microsoft_fabric.map.list_mapsReturns a list of Maps from the specified workspace | Returns a list of Maps from the specified workspace. | read | this connector |
Try it
Prompts to paste into Claude.
Each one names the Microsoft Fabric actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Returns a list of workspaces the principal can access. Use the roles query parameter to filter results by the principal workspace role in Microsoft Fabric and tell me what stands out.
calls
microsoft_fabric.workspaces.list_workspaces - Writewrite
Creates a report in the specified workspace in Microsoft Fabric. Show me exactly what you will send before you save it.
calls
microsoft_fabric.report.create_report
Governance for Microsoft Fabric
Destructive actions wait for a yes.
119 Microsoft Fabric actions are destructive. The first 10:
- Deletes the specified semantic model
microsoft_fabric.semantic_model.delete_semantic_model - Deletes the specified report
microsoft_fabric.report.delete_report - Deletes the specified workspace and the items under it
microsoft_fabric.workspaces.delete_workspace - Create a new execute schedule for a data pipeline
microsoft_fabric.data_pipeline.schedule_execute - Runs an on-demand execute job instance
microsoft_fabric.data_pipeline.run_on_demand_execute - Adds a gateway role assignment
microsoft_fabric.gateways.add_gateway_role_assignment - Adds a workspace role assignment
microsoft_fabric.workspaces.add_workspace_role_assignment - Adds a connection role assignment
microsoft_fabric.connections.add_connection_role_assignment - Adds a deployment pipeline role assignment
microsoft_fabric.deployment_pipelines.add_deployment_pipeline_role_assignment - Updates the specified execute schedule for a data pipeline
microsoft_fabric.data_pipeline.update_execute_schedule
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.
Deletes the specified semantic model?
intent: the user asked me to deletes the specified semantic model
account { op: "keys.create", name: "microsoft-fabric-agent", allowed_actions: ["microsoft_fabric.*"], cap_calls: 2000, confirm_policy: "human_for_writes", approval_rules: [ { match: "microsoft_fabric.semantic_model.delete_semantic_model", approver: "[email protected]" }, { match: "microsoft_fabric.report.delete_report", approver: "[email protected]" }, { match: "microsoft_fabric.workspaces.delete_workspace", approver: "[email protected]" } ]}Cost
What Microsoft Fabric costs on actionlake.
Every Microsoft Fabric action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Microsoft Fabric 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 762 Microsoft Fabric actions
- 133actions pageeach page of a long list is one call
- 278reads 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 Fabric action.
762 actions by area: the id your agent calls, what it does, and its effect.
Workspaces35
microsoft_fabric.workspaces.create_workspaceCreates a new workspacewritemicrosoft_fabric.workspaces.get_workspaceReturns specified workspace informationread- and 33 more in this area
Data agent31
microsoft_fabric.data_agent.list_data_agentsReturns a list of DataAgents from the specified workspacereadmicrosoft_fabric.data_agent.list_datasource_elementsReturns schema elements at a level of the schema tree for a staging datasourceread- and 29 more in this area
Apache airflow job24
microsoft_fabric.apache_airflow_job.list_airflow_pool_templates_betaList Apache Airflow pool templatesreadmicrosoft_fabric.apache_airflow_job.list_apache_airflow_job_files_betaReturns a list of Apache Airflow job files from the specified Apache Airflow jobread- and 22 more in this area
Environment23
microsoft_fabric.environment.get_environmentReturns properties of the specified environmentreadmicrosoft_fabric.environment.list_environmentsReturns a list of environments from the specified workspaceread- and 21 more in this area
Lakehouse23
microsoft_fabric.lakehouse.list_lakehousesReturns a list of lakehouses from the specified workspacereadmicrosoft_fabric.lakehouse.list_livy_sessionsReturns a list of livy sessions from the specified item identifierread- and 21 more in this area
Admin domains19
microsoft_fabric.admin_domains.get_domainReturns the specified domain inforeadmicrosoft_fabric.admin_domains.get_domain_previewReturns the specified domain info (preview)read- and 17 more in this area
Eventstream18
microsoft_fabric.eventstream.get_eventstreamReturns properties of the specified eventstreamreadmicrosoft_fabric.eventstream.get_eventstream_destinationReturns the specified destination of the eventstreamread- and 16 more in this area
Deployment pipelines17
microsoft_fabric.deployment_pipelines.list_deployment_pipeline_operationsReturns a list of the up-to-20 most recent deploy operations performed on the specified deployment pipelinereadmicrosoft_fabric.deployment_pipelines.list_deployment_pipeline_role_assignmentsReturns a list of deployment pipeline role assignmentsread- and 15 more in this area
Gateways17
microsoft_fabric.gateways.get_gatewayGet gateway by IDreadmicrosoft_fabric.gateways.list_gateway_membersLists gateway members of an OnPremisesGateway by IDread- and 15 more in this area
Warehouse17
microsoft_fabric.warehouse.get_connection_stringReturns the SQL connection string of the specified warehousereadmicrosoft_fabric.warehouse.get_restore_pointReturns the properties of a restore point specified for a warehouseread- and 15 more in this area
Items16
microsoft_fabric.items.get_downstream_relations_betaGets downstream relations for an itemreadmicrosoft_fabric.items.get_itemReturns properties of the specified itemread- and 14 more in this area
Policy set16
microsoft_fabric.policy_set.get_active_for_capacityReturns the active policy set on the specified capacityreadmicrosoft_fabric.policy_set.get_policy_rule_betaReturns a single policy rule from the specified policy setread- and 14 more in this area
Spark16
microsoft_fabric.spark.list_capacity_custom_pools_betaList custom poolsreadmicrosoft_fabric.spark.list_livy_sessionsReturns a list of livy sessions from the specified workspaceread- and 14 more in this area
Data pipeline15
microsoft_fabric.data_pipeline.get_data_pipelineReturns properties of the specified data pipelinereadmicrosoft_fabric.data_pipeline.list_data_pipelinesReturns a list of data pipelines from the specified workspaceread- and 13 more in this area
Dataflow15
microsoft_fabric.dataflow.discover_dataflow_parametersRetrieves all parameters defined in the specified Dataflowreadmicrosoft_fabric.dataflow.get_dataflowReturns properties of the specified Dataflowread- and 13 more in this area
ML model15
microsoft_fabric.ml_model.get_ml_modelReturns properties of the specified machine learning modelreadmicrosoft_fabric.ml_model.get_ml_model_endpointReturns properties of the specified machine learning model's endpointread- and 13 more in this area
Git14
microsoft_fabric.git.get_connectionReturns git connection details for the specified workspacereadmicrosoft_fabric.git.get_connection_settingsGets the connection settings for a Git-connected workspaceread- and 12 more in this area
SQL database13
microsoft_fabric.sql_database.get_sql_audit_settingsGets the auditing settings on the specified SQL databasereadmicrosoft_fabric.sql_database.get_sql_databaseReturns properties of the specified SQL databaseread- and 11 more in this area
Connections12
microsoft_fabric.connections.get_connectionGet connection by IDreadmicrosoft_fabric.connections.list_connection_role_assignmentsReturns a list of connection role assignmentsread- and 10 more in this area
Mirrored catalog12
microsoft_fabric.mirrored_catalog.list_mirrored_catalogsReturns a list of MirroredCatalogs from the specified workspacereadmicrosoft_fabric.mirrored_catalog.list_scopes_betaReturns a hierarchical list of namespaces from the source catalogread- and 10 more in this area
Notebook12
microsoft_fabric.notebook.get_livy_sessionReturns properties of the specified livy sessionreadmicrosoft_fabric.notebook.list_livy_sessionsReturns a list of livy sessions from the specified item identifierread- and 10 more in this area
KQL database11
microsoft_fabric.kql_database.get_kql_databaseReturns properties of the specified KQL databasereadmicrosoft_fabric.kql_database.get_shortcutReturns the properties of a table shortcut under a databaseread- and 9 more in this area
Mirrored azure databricks catalog11
microsoft_fabric.mirrored_azure_databricks_catalog.discover_catalogsReturns a list of catalogs from Unity Catalogreadmicrosoft_fabric.mirrored_azure_databricks_catalog.discover_schemasReturns a list of schemas in the given catalog from Unity Catalogread- and 9 more in this area
Mirrored database11
microsoft_fabric.mirrored_database.create_mirrored_databaseCreates a mirrored database in the specified workspacewritemicrosoft_fabric.mirrored_database.get_mirrored_databaseReturns properties of the specified mirrored databaseread- and 9 more in this area
Spark job definition11
microsoft_fabric.spark_job_definition.get_livy_sessionReturns properties of the specified livy sessionreadmicrosoft_fabric.spark_job_definition.list_livy_sessionsReturns a list of livy sessions from the specified item identifierread- and 9 more in this area
Copy job10
microsoft_fabric.copy_job.create_copy_jobCreates a CopyJob in the specified workspacewritemicrosoft_fabric.copy_job.get_copy_jobReturns properties of the specified CopyJobread- and 8 more in this area
Graph model10
microsoft_fabric.graph_model.create_graph_modelCreates a GraphModel in the specified workspacewritemicrosoft_fabric.graph_model.get_graph_modelReturns properties of the specified GraphModelread- and 8 more in this area
Data build tool job9
microsoft_fabric.data_build_tool_job.create_data_build_tool_jobCreates a DataBuildToolJob in the specified workspacewritemicrosoft_fabric.data_build_tool_job.get_data_build_tool_jobReturns properties of the specified DataBuildToolJobread- and 7 more in this area
Job scheduler9
microsoft_fabric.job_scheduler.get_item_job_instanceGet one item's job instancereadmicrosoft_fabric.job_scheduler.get_item_scheduleGet an existing schedule for an itemread- and 7 more in this area
Monitoring9
microsoft_fabric.monitoring.create_monitoringCreates a Monitoring in the specified workspacewritemicrosoft_fabric.monitoring.get_monitoringGets the specified Monitoringread- and 7 more in this area
Admin tenants 8Admin workspaces 8Reflex 8Semantic model 8Anomaly detector 7Azure databricks storage 7Cosmos DB database 7Digital twin builder 7Digital twin builder flow 7Event schema set 7Eventhouse 7Graph query set 7Graphql API 7KQL dashboard 7KQL queryset 7Map 7Mounted data factory 7Ontology 7Operations agent 7Org app 7Org app audience 7Paginated report 7Plan 7Report 7Snowflake database 7User data function 7Variable library 7Folders 6One lake shortcuts 6SQL endpoint 6App backend 5External data shares provider 5ML experiment 5One lake data access security 5One lake settings 5Warehouse snapshot 5Admin tags 4Admin workloads 4Capacities 4Managed private endpoints 4Admin items 3Admin policy sets 3Recoverable items 3Tags 3Admin external data shares provider 2Admin labels 2Admin sharing links 2Domains 2External data shares recipient 2Long running operations 2One lake lifecycle policy 2Surge protection 2Admin users 1Catalog 1Dashboard 1Datamart 1Mirrored warehouse 1Real time intelligence 1
This index shows a sample of each area. All 762 actions, with their inputs and prices, are in the catalogue, which a free sandbox key can search.
Questions
The Microsoft Fabric MCP server, answered.
Is there a Microsoft Fabric MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 762 Microsoft Fabric actions across workspaces, data agent, apache airflow job and environment, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Microsoft Fabric.
What do I need to connect Microsoft Fabric?
Entra application (client) id of an app registration allowed to call Fabric APIs and Entra client secret. Plus your tenant id. The free call account { op: "connections.requirements", service: "microsoft_fabric" } lists exactly what to paste.
Can an AI agent delete or change things in Microsoft Fabric?
Only if its key allows it, and only with an explicit confirmation. 119 Microsoft Fabric actions are destructive, including deletes the specified semantic model, deletes the specified report and deletes the specified workspace and the items under it. 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 Fabric API limit?
actionlake paces every call to the limits in the Microsoft Fabric service file: 100 calls a minute per connection. Repeated reads are answered from cache (278 Microsoft Fabric actions can be), which never reaches Microsoft Fabric.
What does a Microsoft Fabric action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (133 Microsoft Fabric actions page). Cache hits, searches, dry runs and calls Microsoft Fabric 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 Fabric API does it cover?
The Fabric REST API v1 (fabric-rest-api-specs, Swagger 2.0) at api.fabric.microsoft.com/v1. It is built from Microsoft Fabric's documentation as read on 22 October 2026. 762 actions in 88 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Microsoft Fabric 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 Fabric 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 Fabric today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.
