Package org.mockserver.model

Class LlmContentFilter

java.lang.Object

org.mockserver.model.ObjectWithReflectiveEqualsHashCodeToString

org.mockserver.model.ObjectWithJsonToString

org.mockserver.model.LlmContentFilter

public class LlmContentFilter
extends ObjectWithJsonToString

Azure OpenAI content-filter severities for the four core harm categories (hate, sexual, violence, self_harm). Azure annotates every completion with these results — content_filter_results on each choice and prompt_filter_results at the top level — which agents read to detect when a response was filtered.

Each category carries a severity of safe, low, medium or high (default safe). Azure's default policy filters content at medium or high, so the derived filtered flag is true at those severities. Used both to annotate a normal Azure response and, on the chaos path, to shape a content-filter block.

Additive/opt-in — an absent LlmContentFilter leaves Azure responses unchanged. Follows the model field/withX/getter convention so it round-trips through the schema-validated expectation JSON.

Field Summary

Fields

Modifier and Type	Field	Description
static final String	HIGH
static final String	LOW
static final String	MEDIUM
static final String	SAFE

Constructor Summary

Constructors

Constructor	Description
LlmContentFilter()

Method Summary

Modifier and Type	Method	Description
boolean	equals(Object o)
String	getHate()
String	getSelfHarm()
String	getSexual()
String	getViolence()
int	hashCode()
static boolean	isFilteredSeverity(String severity)	Azure filters at medium or high; everything else is delivered.
static LlmContentFilter	llmContentFilter()
static String	severityOrSafe(String severity)	Severity for a category, defaulting to safe when unset.
LlmContentFilter	withHate(String hate)
LlmContentFilter	withSelfHarm(String selfHarm)
LlmContentFilter	withSexual(String sexual)
LlmContentFilter	withViolence(String violence)

Methods inherited from class org.mockserver.model.ObjectWithJsonToString

toString

Methods inherited from class org.mockserver.model.ObjectWithReflectiveEqualsHashCodeToString

fieldsExcludedFromEqualsAndHashCode

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Field Details

SAFE

public static final String SAFE

See Also:

• Constant Field Values

LOW

public static final String LOW

See Also:

• Constant Field Values

MEDIUM

public static final String MEDIUM

See Also:

• Constant Field Values

HIGH

public static final String HIGH

See Also:

• Constant Field Values

Constructor Details

LlmContentFilter

public LlmContentFilter()

Method Details

llmContentFilter

public static LlmContentFilter llmContentFilter()

withHate

public LlmContentFilter withHate(String hate)

getHate

public String getHate()

withSexual

public LlmContentFilter withSexual(String sexual)

getSexual

public String getSexual()

withViolence

public LlmContentFilter withViolence(String violence)

getViolence

public String getViolence()

withSelfHarm

public LlmContentFilter withSelfHarm(String selfHarm)

getSelfHarm

public String getSelfHarm()

severityOrSafe

public static String severityOrSafe(String severity)

Severity for a category, defaulting to safe when unset.

isFilteredSeverity

public static boolean isFilteredSeverity(String severity)

Azure filters at medium or high; everything else is delivered.

equals

public boolean equals(Object o)

Overrides:

equals in class ObjectWithReflectiveEqualsHashCodeToString

hashCode

public int hashCode()

Overrides:

hashCode in class ObjectWithReflectiveEqualsHashCodeToString