agentleFS
Microsoft AzureKnown publisherAzure/connectors-python-sdk/.github/skills/connection-setup/SKILL.md

connection-setup

Skill0 starsChanged 18 days agoMITView on GitHub

Create and configure Connector Namespace connections for the Azure Connectors Python SDK. USE WHEN: setting up a new connector connection, creating a Connector Namespace, authorizing OAuth consent, adding access policies, or configuring local.settings.json / app settings. Covers Office365, SharePoint, Teams, and any Microsoft.Web/connections connector. Works with Azure Functions, Flask, FastAPI, Django, or any Python app using the SDK. NOT FOR: trigger registration (use trigger-registration skill), or code generation.

4 other projects use a version of this, 4 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 2 are common, 2 are distinctive, and 0 are very specific to this project.

  • 2

    Common

    Most of the 4 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Common:

    Check for an existing Connector Namespace in the resource group:

    line 28
  2. Common:

    If none exists, create one:

    line 39
  3. Distinctive:

    Test the connection works end-to-end:

    line 163
  4. Distinctive:

    Use the connection runtime URL with a caller-owned async Azure credential.

    line 183

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Connector Namespace Connection Setup

Automates the end-to-end connection lifecycle for SDK-supported connectors. Connections created here can be used by any Python app via the azure-connectors package — Azure Functions, Flask, FastAPI, Django, scripts, etc.

When to Use

  • Developer needs a new connector connection for local dev or a deployed compute host
  • Developer needs to authorize (OAuth consent) a connection
  • Developer needs to wire connection URLs into local.settings.json or deployed app settings
  • Developer needs to grant access policies (CLI identity for local, managed identity for deployed)

Prerequisites

  • Azure CLI installed and authenticated (az login)
  • Target subscription and resource group known
  • For deployed scenarios: compute host (e.g., Function App, App Service) with managed identity enabled
  • Supported regions for Connector Namespace: brazilsouth, centraluseuap, eastus2euap, centralusstage, eastusstage. Only the Connector Namespace location must be in a supported region; the resource group and Function App can be in any region.

Procedure

Step 1: Create or Select Connector Namespace

Check for an existing Connector Namespace in the resource group:

$subscriptionId = "<subscription-id>"
$resourceGroup = "<resource-group>"

az rest --method GET `
    --uri "https://management.azure.com/subscriptions/$subscriptionId/resourceGroups/$resourceGroup/providers/Microsoft.Web/connectorGateways?api-version=2026-05-01-preview" `
    -o json | ConvertFrom-Json | Select-Object -ExpandProperty value | Select-Object name

If none exists, create one:

$connectorNamespace= "<connector-namespace>"
$location = "<azure-region>"

$gwBody = "{`"location`":`"$location`",`"identity`":{`"type`":`"SystemAssigned`"},`"properties`":{}}"
$tempFile = Join-Path $env:TEMP "gw-body.json"
[System.IO.File]::WriteAllText($tempFile, $gwBody)
az rest --method PUT `
    --uri "https://management.azure.com/subscriptions/$subscriptionId/resourceGroups/$resourceGroup/providers/Microsoft.Web/connectorGateways/$connectorNamespace?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json
Remove-Item $tempFile -ErrorAction SilentlyContinue

Important: The Connector Namespace must have a managed identity enabled (SystemAssigned) for trigger callback authentication. If the Connector Namespace was created without an identity, update it:

$gwBody = "{`"location`":`"$location`",`"identity`":{`"type`":`"SystemAssigned`"},`"properties`":{}}"
$tempFile = Join-Path $env:TEMP "gw-identity.json"
[System.IO.File]::WriteAllText($tempFile, $gwBody)
az rest --method PUT `
    --uri "https://management.azure.com/subscriptions/$subscriptionId/resourceGroups/$resourceGroup/providers/Microsoft.Web/connectorGateways/$connectorNamespace?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json
Remove-Item $tempFile -ErrorAction SilentlyContinue

Step 2: Create Connection

Supported SDK connector names: arm, azuread, azureautomation, azureblob, azuredatafactory, azuredigitaltwins, azureeventgrid, azureiotcentral, azuremonitorlogs, azurequeues, azuretables, azurevm, box, campfire, clicksendsms, cloudconvert, cloudmersiveconvert, commondataservice, dataflows, documentdb, docusign, docuware, dropbox, dynamicsax, elfsquaddata, etsy, eventbrite, eventhubs, excelonline, excelonlinebusiness, formstackforms, freshservice, ftp, github, googlecalendar, googledrive, googletasks, impexium, infusionsoft, insightly, jedoxodatahub, jira, keyvault, kusto, mailchimp, meetingroommap, microsoftbookings, microsoftforms, monday, mq, msgraphgroupsanduser, office365, office365groups, office365groupsmail, office365users, onedrive, onedriveforbusiness, onenote, orderful, outlook, pdfco, pipedrive, planner, plivo, plumsail, powerbi, projectplace, replicon, revai, rss, salesforce, seismicplanner, sendgrid, servicebus, sharepointonline, shifts, signinghub, slack, smtp, sql, starmind, starrezrestv1, tallyfy, teams, textrequest, ticketmaster, todo, trello, twitter, typeform, universalprint, visualstudioteamservices, waywedo, wdatp, webex, wordonlinebusiness, wordpress, yammer, zendesk, zeptomail, zohosign (and any Microsoft.Web/connections connector name).

$connectorName = "<connector-name>"      # e.g., "azureblob", "kusto", "mq", "msgraphgroupsanduser", "office365", "office365users", "sharepointonline", "teams"
$connectionName = "<connection-name>"    # e.g., "office365-test", "sharepoint-test"

$gwId = "/subscriptions/$subscriptionId/resourceGroups/$resourceGroup/providers/Microsoft.Web/connectorGateways/$connectorNamespace"
$connBody = "{`"properties`":{`"connectorName`":`"$connectorName`"}}"
$tempFile = Join-Path $env:TEMP "conn-body.json"
[System.IO.File]::WriteAllText($tempFile, $connBody)
az rest --method PUT `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json | ConvertFrom-Json | Select-Object name, @{n='status';e={$_.properties.statuses[0].status}}
Remove-Item $tempFile -ErrorAction SilentlyContinue

The connection starts in Error state (unauthenticated). Proceed to Step 3.

Step 3: OAuth Consent (In-Browser)

Retrieve the consent link and open it in the default browser — no portal needed:

$consentBody = '{"parameters":[{"redirectUrl":"https://portal.azure.com","parameterName":"token"}]}'
$tempFile = Join-Path $env:TEMP "consent-body.json"
[System.IO.File]::WriteAllText($tempFile, $consentBody)
$result = az rest --method POST `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}/listConsentLinks?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json | ConvertFrom-Json
Remove-Item $tempFile -ErrorAction SilentlyContinue

$link = $result.value[0].link
Start-Process $link

The user completes OAuth in the browser. After consent, verify:

az rest --method GET `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}?api-version=2026-05-01-preview" `
    -o json | ConvertFrom-Json | Select-Object @{n='status';e={$_.properties.statuses[0].status}}

Expected: Connected.

Step 4: Get Connection Runtime URL

$conn = az rest --method GET `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}?api-version=2026-05-01-preview" `
    -o json | ConvertFrom-Json
$runtimeUrl = $conn.properties.connectionRuntimeUrl
Write-Output "Runtime URL: $runtimeUrl"

Step 5: Add Access Policies

Note: Access policies control which identities can call the connection's runtime URL for connector actions (e.g., send email, list files). For trigger-only scenarios, the Connector Namespace polls server-side and does not need an access policy on the connection. Skip this step if your function only receives trigger callbacks and does not call connector actions at runtime.

For local development (Azure CLI identity)

$userObjectId = az ad signed-in-user show --query "id" -o tsv
$tenantId = az account show --query "tenantId" -o tsv

$policyBody = "{`"properties`":{`"principal`":{`"type`":`"ActiveDirectory`",`"identity`":{`"objectId`":`"$userObjectId`",`"tenantId`":`"$tenantId`"}}}}"
$tempFile = Join-Path $env:TEMP "policy-body.json"
[System.IO.File]::WriteAllText($tempFile, $policyBody)
az rest --method PUT `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}/accessPolicies/local-dev?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json | ConvertFrom-Json | Select-Object name
Remove-Item $tempFile -ErrorAction SilentlyContinue

For deployed compute host (e.g., Function App with system-assigned MSI)

$functionAppName = "<function-app-name>"
$msiObjectId = az functionapp identity show -g $resourceGroup -n $functionAppName --query "principalId" -o tsv
$tenantId = az account show --query "tenantId" -o tsv

$policyBody = "{`"properties`":{`"principal`":{`"type`":`"ActiveDirectory`",`"identity`":{`"objectId`":`"$msiObjectId`",`"tenantId`":`"$tenantId`"}}}}"
$tempFile = Join-Path $env:TEMP "msi-policy-body.json"
[System.IO.File]::WriteAllText($tempFile, $policyBody)
az rest --method PUT `
    --uri "https://management.azure.com${gwId}/connections/${connectionName}/accessPolicies/functionapp-msi?api-version=2026-05-01-preview" `
    --body "@$tempFile" --headers "Content-Type=application/json" -o json | ConvertFrom-Json | Select-Object name
Remove-Item $tempFile -ErrorAction SilentlyContinue

ACL propagation takes 1-5 minutes. If you get 403 errors immediately after adding, wait and retry.

Step 6: Verify Connection

Test the connection works end-to-end:

# Office365
az rest --method GET --uri "$runtimeUrl/Categories" --resource "https://apihub.azure.com" -o json

# SharePoint
az rest --method GET --uri "$runtimeUrl/datasets" --resource "https://apihub.azure.com" -o json

# Teams — list joined teams to verify Teams connection
az rest --method GET --uri "$runtimeUrl/beta/me/joinedTeams" --resource "https://apihub.azure.com" -o json

# Azure Blob — list datasets
az rest --method GET --uri "$runtimeUrl/v2/datasets" --resource "https://apihub.azure.com" -o json

Next Steps

SDK Usage

Use the connection runtime URL with a caller-owned async Azure credential. Root clients are async-only:

from azure.identity.aio import DefaultAzureCredential

from azure.connectors.office365 import Office365Client, SendEmailInput

async with DefaultAzureCredential() as credential:
    async with Office365Client(runtime_url, credential) as client:
        await client.send_email(
            input=SendEmailInput(
                to="recipient@example.com",
                subject="Hello",
                body="Sent with Azure Connectors.",
            )
        )

Clients and models share the root connector module. Use azure.identity.aio credentials and await unsuffixed operations. No synchronous mirror is provided.

  • Azure Functions triggers: To register connector triggers (e.g., OnNewEmail, OnNewFile) for Azure Functions, use the trigger-registration skill.
  • Azure Functions signatures: For a complete mapping of trigger operations to Azure Functions signatures, see Operations to Functions Signature Match.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.