WolfDefender class abstract

Detects prompt injections and jailbreaks in text with Patronus's Wolf Defender Edge (80M parameters, Apache 2.0): for example in email, documents, web pages, and tool results before an AI model reads them.

The text has no length limit. Each 2046 tokens is one model run, and the results combine. The model was trained mainly on English and German.

Implementers

• WolfDefenderNative
• WolfDefenderWeb

Constructors

WolfDefender()

Properties

bundle → WolfDefenderBundle

no setter

hashCode → int

The hash code for this object.

no setter, inherited

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

classify(String text) → Future<WolfDefenderResult>

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited

Static Methods

load(WolfDefenderBundle bundle) → WolfDefender

Returns the one instance. It loads the model on first use.