microsoft/azure-mgmt-fabric-py
|- Azure Fabric Management SDK for Python. Use for managing Microsoft Fabric capacities and resources.
npx skills add https://github.com/microsoft/skills --skill azure-mgmt-fabric-py
Manage Microsoft Fabric capacities and resources programmatically.
pip install azure-mgmt-fabric
pip install azure-identity
AZURE_SUBSCRIPTION_ID=<your-subscription-id> # Required for all auth methods
AZURE_RESOURCE_GROUP=<your-resource-group> # Required for all auth methods
AZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production
> 🔑 Two rules apply to every code sample below:
>
> 1. Prefer DefaultAzureCredential. It works locally (Azure CLI / VS Code / Developer CLI) and in Azure (managed identity, workload identity) with no code change. Avoid connection strings, account/API keys — they bypass Entra audit and rotation.
> - Local dev: DefaultAzureCredential works as-is.
> - Production: set AZURE_TOKEN_CREDENTIALS=prod (or AZURE_TOKEN_CREDENTIALS=<specific_credential>) to constrain the credential chain to production-safe credentials.
> 2. Wrap every client in a context manager so HTTP transports, sockets, and token caches are released deterministically:
> - Sync: with <Client>(...) as client:
> - Async: async with <Client>(...) as client: and async with DefaultAzureCredential() as credential: (from azure.identity.aio)
>
> Snippets may abbreviate this setup, but production code should always follow both rules.
from azure.identity import DefaultAzureCredential, ManagedIdentityCredential
from azure.mgmt.fabric import FabricMgmtClient
import os
# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=<specific_credential>
credential = DefaultAzureCredential(require_envvar=True)
# Or use a specific credential directly in production:
# See https://learn.microsoft.com/python/api/overview/azure/identity-readme?view=azure-python#credential-classes
# credential = ManagedIdentityCredential()
with FabricMgmtClient(
credential=credential,
subscription_id=os.environ["AZURE_SUBSCRIPTION_ID"]
) as client:
# Use `client` for all subsequent operations (see examples below)
...
from azure.mgmt.fabric import FabricMgmtClient
from azure.mgmt.fabric.models import FabricCapacity, FabricCapacityProperties, CapacitySku
from azure.identity import DefaultAzureCredential
import os
resource_group = os.environ["AZURE_RESOURCE_GROUP"]
capacity_name = "myfabriccapacity"
credential = DefaultAzureCredential()
with FabricMgmtClient(
credential=credential,
subscription_id=os.environ["AZURE_SUBSCRIPTION_ID"]
) as client:
capacity = client.fabric_capacities.begin_create_or_update(
resource_group_name=resource_group,
capacity_name=capacity_name,
resource=FabricCapacity(
location="eastus",
sku=CapacitySku(
name="F2", # Fabric SKU
tier="Fabric"
),
properties=FabricCapacityProperties(
administration=FabricCapacityAdministration(
members=["[email protected]"]
)
)
)
).result()
print(f"Capacity created: {capacity.name}")
capacity = client.fabric_capacities.get(
resource_group_name=resource_group,
capacity_name=capacity_name
)
print(f"Capacity: {capacity.name}")
print(f"SKU: {capacity.sku.name}")
print(f"State: {capacity.properties.state}")
print(f"Location: {capacity.location}")
capacities = client.fabric_capacities.list_by_resource_group(
resource_group_name=resource_group
)
for capacity in capacities:
print(f"Capacity: {capacity.name} - SKU: {capacity.sku.name}")
all_capacities = client.fabric_capacities.list_by_subscription()
for capacity in all_capacities:
print(f"Capacity: {capacity.name} in {capacity.location}")
from azure.mgmt.fabric.models import FabricCapacityUpdate, CapacitySku
updated = client.fabric_capacities.begin_update(
resource_group_name=resource_group,
capacity_name=capacity_name,
properties=FabricCapacityUpdate(
sku=CapacitySku(
name="F4", # Scale up
tier="Fabric"
),
tags={"environment": "production"}
)
).result()
print(f"Updated SKU: {updated.sku.name}")
Pause capacity to stop billing:
client.fabric_capacities.begin_suspend(
resource_group_name=resource_group,
capacity_name=capacity_name
).result()
print("Capacity suspended")
Resume a paused capacity:
client.fabric_capacities.begin_resume(
resource_group_name=resource_group,
capacity_name=capacity_name
).result()
print("Capacity resumed")
client.fabric_capacities.begin_delete(
resource_group_name=resource_group,
capacity_name=capacity_name
).result()
print("Capacity deleted")
from azure.mgmt.fabric.models import CheckNameAvailabilityRequest
result = client.fabric_capacities.check_name_availability(
location="eastus",
body=CheckNameAvailabilityRequest(
name="my-new-capacity",
type="Microsoft.Fabric/capacities"
)
)
if result.name_available:
print("Name is available")
else:
print(f"Name not available: {result.reason}")
skus = client.fabric_capacities.list_skus(
resource_group_name=resource_group,
capacity_name=capacity_name
)
for sku in skus:
print(f"SKU: {sku.name} - Tier: {sku.tier}")
| Operation | Method |
|-----------|--------|
| client.fabric_capacities | Capacity CRUD operations |
| client.operations | List available operations |
| SKU | Description | CUs |
|-----|-------------|-----|
| F2 | Entry level | 2 Capacity Units |
| F4 | Small | 4 Capacity Units |
| F8 | Medium | 8 Capacity Units |
| F16 | Large | 16 Capacity Units |
| F32 | X-Large | 32 Capacity Units |
| F64 | 2X-Large | 64 Capacity Units |
| F128 | 4X-Large | 128 Capacity Units |
| F256 | 8X-Large | 256 Capacity Units |
| F512 | 16X-Large | 512 Capacity Units |
| F1024 | 32X-Large | 1024 Capacity Units |
| F2048 | 64X-Large | 2048 Capacity Units |
| State | Description |
|-------|-------------|
| Active | Capacity is running |
| Paused | Capacity is suspended (no billing) |
| Provisioning | Being created |
| Updating | Being modified |
| Deleting | Being removed |
| Failed | Operation failed |
All mutating operations are long-running (LRO). Use .result() to wait:
# Synchronous wait
capacity = client.fabric_capacities.begin_create_or_update(...).result()
# Or poll manually
poller = client.fabric_capacities.begin_create_or_update(...)
while not poller.done():
print(f"Status: {poller.status()}")
time.sleep(5)
capacity = poller.result()
azure.xxx sync clients with azure.xxx.aio async clients in the same call path. Choose one mode per module.with Client(...) as client: (sync) or async with Client(...) as client: (async). For async DefaultAzureCredential from azure.identity.aio, also use async with credential: so tokens and transports are cleaned up.DefaultAzureCredential for code that runs locally. Use a specific token credential for code that runs in Azure.10. Monitor capacity usage via Azure Monitor metrics
| File | Contents |
|------|----------|
| references/capabilities.md | Additional non-hero capabilities, operation-group coverage, and production checklists. |
| references/non-hero-scenarios.md | Dedicated non-hero examples for secondary/advanced scenarios. |
Take microsoft/azure-mgmt-fabric-py from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.
The instructions reference pip.
Without those the skill loads but fails at the first command.