Index of every page on this site: llms.txt
All skills
microsoft avatar

/azure-ai

@0d60658
by microsoftmicrosoft/skills3.2k stars
373
Run it yourself

Build and troubleshoot Azure AI Search, Speech, Document Intelligence, Content Understanding, Translator, Content Safety, and OpenAI apps. WHEN: AI Search, vector search, hybrid search, semantic search, indexer, speech-to-text, text-to-speech, fast or batch transcription, Speech SDK logs, session ID, result ID, OCR, document translation, glossary.

Use this Skill: https://skilld.dev/gh/microsoft/skills/azure-ai

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencescontent-understandingdiagnostics.md

≈1k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Content Understanding diagnostics

For failing Content Understanding (CU) calls, Studio pages that fail, results that look wrong, or analyses that never finish. Learn has no dedicated CU error-code page: errors use the Azure Core error shape plus an x-ms-error-code header, so branch on error.code, nested innererror.code and the header, never on message text.

Workflow

  1. Record UTC time, api-version, operation (create, analyze, analyzeBinary, GET result, defaults), HTTP status and error codes, and the result ID from Operation-Location (evidence without content).
  2. Match the symptom:
Symptom Check first
401/403 on /contentunderstanding/*, Foundry portal CU page or CU Studio project list Role at the resource scope, auth mode and disableLocalAuth (access and roles; run the access checker)
403 with a valid role, timeouts, Studio loads but calls fail Network rules, private endpoint DNS for each host (private networking)
Error about a missing default or model deployment GET .../contentunderstanding/defaults, deployment names, the analyzer's supportedModels, or pass modelDeployments per request (overview and setup)
URL input cannot be fetched The URL must point to one object the CU resource can reach: SAS validity, read permission, storage firewall; or send bytes with analyzeBinary (Studio and storage)
A property or operation is rejected Preview-only feature on GA 2025-11-01; move to 2026-06-01-preview or remove it, and keep the same version on polls
404 on a result ID Wrong ID, wrong endpoint or version, or older than the up-to-24-hour retention; analyze again
HTTP 200/202 but status Running for long or Failed Job state is in the body; capture the body status and error code and poll with backoff (every 1–2 s per Learn samples)
Succeeded but fields missing result.warnings guardrail codes, field method, description quality, labeled samples (custom analyzers)
Slow or throttled generative steps Quota and metrics of your model deployment (429s there surface in CU)
Studio cannot find the resource Reader to discover the resource, CU role on it, region support, browser network path
  1. Confirm with a minimal plain HTTP call using the same identity, endpoint host and api-version as the app; record status and error codes only.

Read-only Azure checks

Check Azure MCP (if enabled) Without MCP
Model deployment metrics (throttling) monitor_metrics_definitions, then monitor_metrics_query on the resource az monitor metrics list --resource <resource-id> --metric <name> --interval PT5M after az monitor metrics list-definitions --resource <resource-id>
Configuration changes monitor_activitylog_list az monitor activity-log list --resource-id <resource-id> --offset 7d
Service health resourcehealth_availability-status_get Azure portal Resource Health / Service Health

No incident or health event is not proof that the service is healthy, and a successful test does not validate the app's identity, network or document.

Sources (retrieved 2026-11-18): CU REST quickstart · Data privacy (retention) · Models and deployments · Guardrails · Migrate preview to GA

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning11d4 checks · Risk SAFE
  • Gen Agent Trust Hub11d

    This skill provides tools and guidance for developing and troubleshooting Azure AI services. It includes a local diagnostic script for analyzing Speech SDK logs with robust safety controls, along with comprehensive references for official Azure SDKs and secure authentication practices.

  • Socket11d

    No alerts

  • Snyk11d

    Risk: LOW · No issues

  • Runlayer9mo

    15/15 files flagged

Signed by skilld at 0d60658. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 days ago
metadata
{
  "author": "Microsoft",
  "version": "1.2.6"
}

README badge

README badge for microsoft/skills/azure-ai