Anthropic Plugin

The genkit_anthropic package provides a unified interface to connect with Anthropic’s Claude models through the Anthropic API.

Setup

Installation

dart pub add genkit_anthropic

Configuration

To use this plugin, import it and specify it when you initialize Genkit:

import 'dart:io';
import 'package:genkit/genkit.dart';

import 'package:genkit_anthropic/genkit_anthropic.dart';

void main() async {
  // Initialize Genkit with the Anthropic plugin
  final ai = Genkit(
    plugins: [anthropic(apiKey: Platform.environment['ANTHROPIC_API_KEY']!)],
  );
}

The plugin requires an Anthropic API Key, which you can get from the Anthropic Console.

Usage

Language Models

You can reference Claude models, e.g. claude-sonnet-4-5 or 'claude-sonnet-4-6'.

import 'dart:io';
import 'package:genkit/genkit.dart';
import 'package:genkit_anthropic/genkit_anthropic.dart';

void main() async {
  final ai = Genkit(
    plugins: [anthropic(apiKey: Platform.environment['ANTHROPIC_API_KEY']!)],
  );

  final response = await ai.generate(
    model: anthropic.model('claude-sonnet-4-5'),
    prompt: 'Tell me a joke about a developer.',
  );
  print(response.text);
}

Streaming

The plugin supports streaming responses natively.

final stream = ai.generateStream(
  model: anthropic.model('claude-sonnet-4-5'),
  prompt: 'Count to 5',
);

await for (final chunk in stream) {
  print(chunk.text);
}

final response = await stream.onResult;
print('Full response: \${response.text}');

Tool Calling

Claude models support function calling and tool use.

import 'package:schemantic/schemantic.dart';

// part 'main.g.dart'; // generated by build_runner

@Schema()
abstract class $CalculatorInput {
  int get a;
  int get b;
}

// ... inside main ...

ai.defineTool(
  name: 'calculator',
  description: 'Multiplies two numbers',
  inputSchema: CalculatorInput.$schema,
  outputSchema: intSchema(),
  fn: (input, context) async => input.a * input.b,
);

final response = await ai.generate(
  model: anthropic.model('claude-sonnet-4-5'),
  prompt: 'What is 123 * 456?',
  toolNames: ['calculator'],
);

print(response.text);

Thinking (Claude 3.7+)

Claude 3.7 models support explicit thinking to expose the reasoning process before returning the final response.

final response = await ai.generate(
  model: anthropic.model('claude-sonnet-4-5'),
  prompt: 'Solve this 24 game: 2, 3, 10, 10',
  config: AnthropicOptions(thinking: ThinkingConfig(budgetTokens: 2048)),
);

// The thinking content is available in the message parts
print(response.message?.content);
print(response.text);

Structured Output

import 'package:schemantic/schemantic.dart';

// part 'main.g.dart'; // generated by build_runner

@Schema()
abstract class $Person {
  String get name;
  int get age;
}

// ... inside main ...

final response = await ai.generate(
  model: anthropic.model('claude-sonnet-4-5'),
  prompt: 'Generate a person named John Doe, age 30',
  outputSchema: Person.$schema,
);

final person = Person.fromJson(response.output!);
print('Name: \${person.name}, Age: \${person.age}');