Content Safety

azure AI serviceazure-ai-content-safety

Detects harmful text and images and screens prompts for jailbreaks, ungrounded answers and protected material.

Provider: azure
Kind: AI service
Category: Compute
Official docs: https://learn.microsoft.com/en-us/azure/ai-services/content-safety/overview

Properties · 7 set by this component

property	type	values	required	kind default	this component
auth	enum	api-key \| oauth \| iam \| none	no	—	▸ api-key
blocklists · component property	boolean	—	no	false	inherited
contentFiltering	boolean	—	no	—	▸ true
customCategories · component property	boolean	—	no	false	inherited
customerManagedKeys	boolean	—	no	—	▸ false
dataResidency · advanced	enum	us \| eu \| asia \| global	no	—	inherited
dataRetentionDays	integer	—	no	—	inherited
keyStore	boolean	—	no	false	inherited
managed	boolean	—	no	false	▸ true
maxImageSizeMb · component property	integer	—	no	4	inherited
maxTextLength · component property	integer	—	no	10000	inherited
modality	enum	text \| vision \| speech \| multimodal \| structured \| image \| video \| audio \| embedding \| rerank	no	—	▸ multimodal
promptShields · component property	boolean	—	no	false	inherited
publicEndpoint	boolean	—	no	—	▸ true
quotaUnitsPerMinute	integer	—	no	—	▸ 6000
rateLimitRpm · advanced	integer	—	no	—	inherited
rateLimitTpm · advanced	integer	—	no	—	inherited
tier · component property	enum	f0 \| s0	no	—	inherited
viaGateway	boolean	—	no	false	inherited

Rules that apply · 2

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

AI API key not managedwarning

ai-api-key-unmanaged

This AI vendor is reached with an API key, and nothing in the design holds or rotates that key. It ends up in environment files or code, every caller shares it, and a single leak runs up the vendor bill until someone notices. Put the key in a key-management component, or say it is held there with keyStore.

— all of these hold:
— the node's `auth` is "api-key"
— the node's `keyStore` is false
— the node has at least one incoming edge from an app-server, worker or client
— this does not hold:
— the design contains a node of kind key-management

Stateful node without zone redundancyerror

stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

— all of these hold:
— the node's `stateless` is false
— the node sits inside exactly 1 availability-zone group
— this does not hold:
— the node sits inside at least one cell group

Use it in a design

nodes:
  - { id: "ai-content-safety", component: "azure-ai-content-safety" }

Other AI service products · 125

and 85 more → AI service page

Source

1# yaml-language-server: $schema=../../schema/component.schema.json
2# Azure AI Content Safety: text and image harm detection plus prompt shields, groundedness and protected material detection.
3# "What is Azure AI Content Safety?" gives the S0 rate limits (1,000 requests per 10 seconds for the moderation APIs, so 6,000 per minute) and the input limits of 10,000 characters for analyze text and 4 MB for analyze image; managed identity is enabled automatically on resource creation.
4# Prompt Shields ("Prompt Shields in Azure AI Content Safety") is its own text:shieldPrompt call that screens user prompts and documents for injection attacks, so promptShields starts off.
5schemaVersion: "1"
6type: "component"
7id: "azure-ai-content-safety"
8kind: "ai-service"
9provider: "azure"
10docs: "https://learn.microsoft.com/en-us/azure/ai-services/content-safety/overview"
11defaults:
12  auth: "api-key"
13  managed: true
14  modality: "multimodal"
15  contentFiltering: true
16  publicEndpoint: true
17  customerManagedKeys: false
18  quotaUnitsPerMinute: 6000
19properties:
20  tier:
21    type: "enum"
22    values: ["f0", "s0"]
23  maxTextLength:
24    type: "integer"
25    default: 10000
26  maxImageSizeMb:
27    type: "integer"
28    default: 4
29  blocklists:
30    type: "boolean"
31    default: false
32  customCategories:
33    type: "boolean"
34    default: false
35  promptShields:
36    type: "boolean"
37    default: false

The same page as markdown, for an agent: /catalogue/components/azure-ai-content-safety.md