Genkit

https://genkit.dev/docs/js
Admin
Genkit is an open-source AI development framework for building full-stack AI-powered applications...
Tokens:158,638
Snippets:1,702
Trust Score:7.6
Update:1 month ago
Show doc for...
Tokens:
Raw
### Start Express Server

Source: https://genkit.dev/docs/js/backend-frameworks/express

Use this command to start the Express server. Ensure you have the necessary dependencies installed.

```bash
npx tsx --watch src/index.ts
```

--------------------------------

### Start Genkit using npm script

Source: https://genkit.dev/docs/js/deployment/firebase

Starts the Genkit development server using a predefined npm script. This is a convenient shortcut for starting the development environment.

```bash
cd $PROJECT_ROOT/functions

npm run genkit:start
```

--------------------------------

### Example package.json for a Genkit Plugin

Source: https://genkit.dev/docs/js/plugin-authoring/overview

This example shows a `package.json` configuration for a Genkit plugin that provides a retriever, embedder, and model. Ensure your package name starts with `genkitx-` and includes relevant Genkit keywords.

```json
{
  "name": "genkitx-my-plugin",
  "keywords": ["genkit-plugin", "genkit-retriever", "genkit-embedder", "genkit-model"],
  // ... dependencies etc.
}
```

--------------------------------

### Start SvelteKit Development Server

Source: https://genkit.dev/docs/js/app-frameworks/sveltekit

Use this command to start the SvelteKit development server. Open the application in your browser to interact with its features.

```bash
npm run dev
```

--------------------------------

### Install Genkit Packages

Source: https://genkit.dev/docs/js/app-frameworks/sveltekit

Install the core Genkit SDK, Google GenAI plugin, fetch wrapper, and CLI tools.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/fetch
npm install -D genkit-cli tsx

```

--------------------------------

### Start Genkit Application

Source: https://genkit.dev/docs/js/evaluation

Command to start your Genkit application, potentially with arguments for your specific app.

```bash
genkit start -- <command to start your app>
```

--------------------------------

### Install Neo4j Plugin

Source: https://genkit.dev/docs/js/integrations/neo4j

Install the Neo4j plugin using npm.

```bash
npm install genkitx-neo4j
```

--------------------------------

### Install Genkit and Related Packages

Source: https://genkit.dev/docs/js/app-frameworks/nuxt

Install the core Genkit SDK, Google GenAI plugin, fetch handlers, and Genkit CLI using npm.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/fetch
npm install -D genkit-cli tsx
```

--------------------------------

### Install Ollama Plugin

Source: https://genkit.dev/docs/js/integrations/ollama

Install the Ollama plugin using npm.

```bash
npm install genkitx-ollama
```

--------------------------------

### Start Genkit Developer UI with NPM Dev

Source: https://genkit.dev/docs/js/devtools

Start the Genkit Developer UI and connect it to a running Node.js development server managed by NPM.

```bash
genkit start -- npm run dev
```

--------------------------------

### Install Azure OpenAI Plugin

Source: https://genkit.dev/docs/js/integrations/azure-foundry

Install the genkitx-azure-openai package using npm or pnpm.

```bash
npm install genkitx-azure-openai
```

--------------------------------

### Install Genkit and Related Packages

Source: https://genkit.dev/docs/js/backend-frameworks/nestjs

Install the core Genkit SDK, Google GenAI plugin, Express handler, and Genkit CLI.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/express
npm install -D genkit-cli
```

--------------------------------

### Add Start and Build Scripts to package.json

Source: https://genkit.dev/docs/js/deployment/any-platform

Define 'start' and 'build' scripts in your package.json file. The 'start' script is used to run the application, and the 'build' script is used to compile TypeScript code.

```json
"scripts": {
  "start": "node --watch lib/index.js",
  "build": "tsc"
}
```

--------------------------------

### Install Chroma Plugin

Source: https://genkit.dev/docs/js/integrations/chroma

Install the ChromaDB plugin for Genkit using npm.

```bash
npm install genkitx-chromadb
```

--------------------------------

### Install Genkit and MCP Plugin

Source: https://genkit.dev/docs/js/model-context-protocol

Install the necessary Genkit and MCP packages using npm.

```bash
npm i genkit @genkit-ai/mcp
```

--------------------------------

### Install Genkit Dependencies

Source: https://genkit.dev/docs/js/tutorials/summarize-youtube-videos

Installs the Genkit core library and the Google GenAI plugin.

```bash
npm install genkit @genkit-ai/google-genai
```

--------------------------------

### Install Genkit CLI with Binary (macOS/Linux)

Source: https://genkit.dev/docs/js/devtools

Install the Genkit CLI on macOS or Linux by downloading and executing a script. This is a preview method.

```bash
curl -sL cli.genkit.dev | bash
```

--------------------------------

### Start Genkit Developer UI and Open in Browser

Source: https://genkit.dev/docs/js/devtools

Start the Genkit Developer UI and automatically open it in your default web browser.

```bash
genkit start -o -- <command to run your code>
```

--------------------------------

### Start Genkit Developer UI with Node Watch

Source: https://genkit.dev/docs/js/devtools

Start the Genkit Developer UI and connect it to a JavaScript file, enabling live reloading on changes.

```bash
genkit start -- node --watch src/index.js
```

--------------------------------

### Start Genkit Developer UI

Source: https://genkit.dev/docs/js/backend-frameworks/express

Launch the Genkit Developer UI alongside your Express application. This command starts the UI and the specified application.

```bash
npx genkit start -- tsx --watch src/index.ts
```

--------------------------------

### Create Next.js App and Install Genkit Client

Source: https://genkit.dev/docs/js/app-frameworks/nextjs

Use this command to create a new Next.js project without an in-project backend and install the Genkit web client.

```bash
npx create-next-app@latest --app --src-dir my-genkit-nextjs
cd my-genkit-nextjs
npm install genkit
```

--------------------------------

### Install Google Cloud Plugin

Source: https://genkit.dev/docs/js/integrations/google-cloud

Install the Google Cloud plugin for Genkit using npm.

```bash
npm install @genkit-ai/google-cloud
```

--------------------------------

### Install Genkit CLI with NPM

Source: https://genkit.dev/docs/js/devtools

Install the Genkit CLI globally using NPM. This is the primary method for most users.

```bash
npm install -g genkit-cli
```

--------------------------------

### Create and Start MCP Server

Source: https://genkit.dev/docs/js/model-context-protocol

Exposes Genkit tools and prompts as an MCP server using `createMcpServer` and starts it with the default stdio transport. Ensure all necessary plugins and definitions are configured before creating the server.

```typescript
import { googleAI } from '@genkit-ai/google-genai';
import { createMcpServer } from '@modelcontextprotocol/sdk/server';
import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';
import { genkit, z } from 'genkit/beta';

const ai = genkit({
  plugins: [googleAI()],
});

ai.defineTool(
  {
    name: 'add',
    description: 'add two numbers together',
    inputSchema: z.object({ a: z.number(), b: z.number() }),
    outputSchema: z.number(),
  },
  async ({ a, b }) => {
    return a + b;
  },
);

ai.definePrompt(
  {
    name: 'happy',
    description: 'everybody together now',
    input: {
      schema: z.object({
        action: z.string().default('clap your hands').optional(),
      }),
    },
  },
  `If you're happy and you know it, {{action}}.`,
);

ai.defineResource(
  {
    name: 'my resouces',
    uri: 'my://resource',
  },
  async () => {
    return {
      content: [
        {
          text: 'my resource',
        },
      ],
    };
  },
);

ai.defineResource(
  {
    name: 'file',
    template: 'file://{path}',
  },
  async ({ uri }) => {
    return {
      content: [
        {
          text: `file contents for ${uri}`,
        },
      ],
    };
  },
);

// Use createMcpServer
const server = createMcpServer(ai, {
  name: 'example_server',
  version: '0.0.1',
});
// Start the server with stdio transport by default
server.start();
```

--------------------------------

### Install Genkit Packages (App Router)

Source: https://genkit.dev/docs/js/app-frameworks/nextjs

Installs the necessary Genkit core, Google GenAI plugin, and Next.js App Router integration packages. Also installs Genkit CLI and tsx for development.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/next
npm install -D genkit-cli tsx
```

--------------------------------

### Install Genkit and Related Packages

Source: https://genkit.dev/docs/js/app-frameworks/angular

Installs the core Genkit SDK, Google GenAI plugin, Express integration, and Genkit CLI for local development and observability.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/express
npm install -D genkit-cli tsx
```

--------------------------------

### Start NestJS Development Server

Source: https://genkit.dev/docs/js/backend-frameworks/nestjs

Use this command to start the NestJS development server. The app will listen on http://localhost:3000 by default.

```bash
npm run start:dev
```

--------------------------------

### Install AWS Bedrock Plugin

Source: https://genkit.dev/docs/js/integrations/aws-bedrock

Install the AWS Bedrock plugin using npm or pnpm.

```bash
npm install genkitx-aws-bedrock
```

--------------------------------

### Install Auth0 AI Genkit

Source: https://genkit.dev/docs/js/integrations/auth0

Install the necessary Auth0 and Genkit packages using npm.

```bash
npm install @auth0/ai @auth0/ai-genkit
```

--------------------------------

### Start Genkit Developer UI

Source: https://genkit.dev/docs/js/app-frameworks/sveltekit

Launch the Genkit Developer UI to test flows and inspect traces. This command starts the UI and watches for changes in your flow files.

```bash
npx genkit start -- tsx --watch src/lib/genkit/bargainChefFlow.ts
```

--------------------------------

### Start Genkit Developer UI

Source: https://genkit.dev/docs/js/app-frameworks/remix

Launch the Genkit Developer UI to test flows and inspect traces. This command starts the UI and watches for changes in your flow files.

```bash
npx genkit start -- tsx --watch app/genkit/bargainChefFlow.ts
```

--------------------------------

### Install Toolbox Client SDK

Source: https://genkit.dev/docs/js/integrations/toolbox

Install the `@toolbox-sdk/core` node package using npm to enable Genkit's use of Toolbox.

```bash
npm install @toolbox-sdk/core
```

--------------------------------

### Install Genkit and Related Packages

Source: https://genkit.dev/docs/js/backend-frameworks/hono

Install necessary Genkit packages for Google GenAI, fetch integration, Hono adapter, and development tools. This includes the core Genkit SDK, Google GenAI plugin, fetch-based server integration, Hono Node.js adapter, and the Genkit CLI with tsx for development.

```bash
npm install genkit @genkit-ai/google-genai @genkit-ai/fetch @hono/node-server
npm install -D genkit-cli tsx
```

--------------------------------

### Install Fastify and Genkit Packages

Source: https://genkit.dev/docs/js/backend-frameworks/fastify

Installs core Fastify, CORS plugin, Genkit SDK, Google GenAI plugin, Fastify integration, and development dependencies including TypeScript and Genkit CLI.

```bash
npm install fastify @fastify/cors genkit @genkit-ai/google-genai @genkit-ai/fastify
npm install -D typescript tsx @types/node genkit-cli
```

--------------------------------

### Start Genkit Developer UI and NestJS Server

Source: https://genkit.dev/docs/js/backend-frameworks/nestjs

Launch the Genkit Developer UI alongside your NestJS development server. This command starts both the UI and the NestJS watch process.

```bash
npx genkit start -- nest start --watch
```

--------------------------------

### Start Genkit with Firebase Emulators

Source: https://genkit.dev/docs/js/deployment/firebase

Starts the Genkit development server alongside the Firebase Local Emulator Suite. This command runs your Genkit flows within the emulated Firebase environment and exposes the Genkit reflection API.

```bash
genkit start -- firebase emulators:start --inspect-functions
```

--------------------------------

### Initialize Firebase Hosting

Source: https://genkit.dev/docs/js/deployment/firebase

Set up Firebase Hosting in your project directory by accepting the default configurations.

```bash
cd $PROJECT_ROOT

firebase init hosting
```

--------------------------------

### Install Genkit Agent Skills

Source: https://genkit.dev/docs/js/backend-frameworks/hono

Install Genkit Agent Skills using npm to get structured guidance on Genkit APIs, patterns, and common errors when coding with an AI assistant.

```bash
npx skills add genkit-ai/skills

```

--------------------------------

### Initiate and Poll Deep Research Operation

Source: https://genkit.dev/docs/js/integrations/google-genai

Start a deep research task using specialized models and poll the operation status until completion. This example uses 'AUTO' visualization and checks the operation every 30 seconds.

```typescript
let { operation } = await ai.generate({
  model: googleAI.model('deep-research-preview-04-2026'),
  prompt: 'Analyze global semiconductor market trends. Include graphics showing market share changes.',
  config: {
    visualization: 'AUTO',
  },
});

if (!operation) throw new Error('No operation returned');

// Deep research operations are long-running and need to be polled
while (!operation.done) {
  operation = await ai.checkOperation(operation);
  await new Promise((resolve) => setTimeout(resolve, 30000)); // Check every 30 seconds
}

console.log(operation.output?.message?.content);
```

--------------------------------

### Create TanStack Start Project

Source: https://genkit.dev/docs/js/app-frameworks/tanstack-start

Use the TanStack CLI to create a new project. Select React with the full-stack option when prompted.

```bash
npx @tanstack/cli@latest create my-genkit-tanstack
cd my-genkit-tanstack
```

--------------------------------

### Toolbox Client Initialization

Source: https://genkit.dev/docs/js/integrations/toolbox

TypeScript example demonstrating how to initialize the Toolbox client with server URL, authentication token, timeout, and retry configurations.

```typescript
import { ToolboxClient } from '@toolbox-sdk/core';

const client = ToolboxClient('https://your-toolbox-server.com', {
  auth: {
    type: 'bearer',
    token: await getAuthToken(), // Your auth token retrieval logic
  },
  timeout: 30000,
  retries: 3,
});
```

--------------------------------

### Create Nuxt Project

Source: https://genkit.dev/docs/js/app-frameworks/nuxt

Use the Nuxt CLI to initialize a new Nuxt project and navigate into the project directory.

```bash
npx nuxi@latest init my-genkit-nuxt
cd my-genkit-nuxt
```

--------------------------------

### Install Anthropic Plugin

Source: https://genkit.dev/docs/js/integrations/anthropic

Install the Anthropic plugin for Genkit using npm.

```bash
npm i --save @genkit-ai/anthropic
```

--------------------------------

### Create Project Directory and Index File

Source: https://genkit.dev/docs/js/tutorials/chat-with-pdf

Sets up the project directory structure and creates the main TypeScript file.

```bash
mkdir -p chat-with-a-pdf/src && \
cd chat-with-a-pdf && \
touch src/index.ts
```

--------------------------------

### Install @cfworker/json-schema Package

Source: https://genkit.dev/docs/js/deployment/any-platform

Install the necessary package for environments that restrict `eval()`.

```bash
npm install @cfworker/json-schema
```

--------------------------------

### Create Fastify Project Structure

Source: https://genkit.dev/docs/js/backend-frameworks/fastify

Sets up a new Fastify project with necessary directories and files for Genkit integration. Initializes npm and TypeScript configuration.

```bash
mkdir my-genkit-fastify
cd my-genkit-fastify
npm init -y
npm pkg set type=module
npx tsc --init
mkdir -p src/genkit
touch src/index.ts src/genkit/bargainChefFlow.ts
```

--------------------------------

### Equivalent TypeScript Interface for Picoschema

Source: https://genkit.dev/docs/js/dotprompt

The TypeScript interface representation of the provided Picoschema example.

```ts
interface Article {
  title: string;
  subtitle?: string | null;
  /** true when in draft state */
  draft?: boolean | null;
  /** approval status */
  status?: 'PENDING' | 'APPROVED' | null;
  /** the date of publication e.g. '2024-04-09' */
  date: string;
  /** relevant tags for article */
  tags: string[];
  authors: {
    name: string;
    email?: string | null;
  }[];
  metadata?: {
    /** ISO timestamp of last update */
    updatedAt?: string | null;
    /** id of approver */
    approvedBy?: number | null;
  } | null;
  /** arbitrary extra data */
  extra?: any;
  /** wildcard field */
  [key: string]: any;
}
```

--------------------------------

### Initialize Node.js Project

Source: https://genkit.dev/docs/js/deployment/any-platform

Initializes a new Node.js project with a package.json file.

```bash
npm init -y
```

--------------------------------

### Add start and build scripts to package.json

Source: https://genkit.dev/docs/js/deployment/cloud-run

Define 'start' and 'build' scripts in your package.json. The 'start' script is essential for Cloud Run to know how to run your application, and 'build' is common for TypeScript projects.

```json
"scripts": {
  "start": "node lib/index.js",
  "build": "tsc"
}
```

--------------------------------

### Install pgvector Dependencies

Source: https://genkit.dev/docs/js/integrations/pgvector

Install the necessary npm packages for PostgreSQL and pgvector integration.

```bash
npm install postgres pgvector
```

--------------------------------

### Install DeepSeek Plugin

Source: https://genkit.dev/docs/js/integrations/deepseek

Install the compatibility package for DeepSeek models using npm.

```bash
npm install @genkit-ai/compat-oai
```

--------------------------------

### Create and Use a Single MCP Client

Source: https://genkit.dev/docs/js/model-context-protocol

Demonstrates how to create a single MCP client instance for a file system server, retrieve its tools, and use them with a Genkit model. Ensure the MCP server is running and accessible.

```typescript
import {
  googleAI,
} from '@genkit-ai/google-genai';
import {
  createMcpClient,
} from '@genkit-ai/mcp';
import { genkit } from 'genkit';

const myFsClient = createMcpClient({
  name: 'myFileSystemClient', // A unique name for this client instance
  mcpServer: {
    command: 'npx',
    args: ['-y', '@modelcontextprotocol/server-filesystem', process.cwd()],
  },
  // rawToolResponses: true, // Optional: get raw MCP responses
});

// In your Genkit configuration:
const ai = genkit({
  plugins: [googleAI()],
});

(async () => {
  await myFsClient.ready();

  // Retrieve tools from this specific client
  const fsTools = await myFsClient.getActiveTools(ai);

  const { text } = await ai.generate({
    model: googleAI.model('gemini-flash-latest'), // Replace with your model
    prompt: 'List files in ' + process.cwd(),
    tools: fsTools,
  });
  console.log(text);

  await myFsClient.disable();
})();

```

--------------------------------

### Quick Start with AWS Bedrock

Source: https://genkit.dev/docs/js/integrations/aws-bedrock

Initialize Genkit with the AWS Bedrock plugin and a specific model for basic text generation.

```typescript
import { genkit } from 'genkit';

import { awsBedrock, amazonNovaProV1 } from 'genkitx-aws-bedrock';

const ai = genkit({
  plugins: [awsBedrock({ region: 'us-east-1' })],
  model: amazonNovaProV1,
});

// Basic usage
const response = await ai.generate({
  prompt: 'What are the key benefits of using AWS Bedrock for AI applications?',
});

console.log(response.text);
```

--------------------------------

### Install Genkit Evaluator Package

Source: https://genkit.dev/docs/js/evaluation

Install the necessary package for the genkitEval plugin to function.

```bash
npm install @genkit-ai/evaluator
```

--------------------------------

### Create SvelteKit Project

Source: https://genkit.dev/docs/js/app-frameworks/sveltekit

Use the SvelteKit CLI to create a new project. Select the minimal template and TypeScript syntax.

```bash
npx sv create my-genkit-sveltekit
cd my-genkit-sveltekit

```

--------------------------------

### Install PDF Parsing Dependencies

Source: https://genkit.dev/docs/js/tutorials/chat-with-pdf

Installs the pdf-parse module for handling PDF files and its TypeScript types.

```bash
npm install pdf-parse && npm install --save-dev @types/pdf-parse
```

--------------------------------

### Install Firebase Plugin

Source: https://genkit.dev/docs/js/integrations/cloud-firestore

Install the Firebase plugin using npm to enable Cloud Firestore integration.

```bash
npm install @genkit-ai/firebase
```

--------------------------------

### Build and Test Node.js Project Locally

Source: https://genkit.dev/docs/js/deployment/any-platform

Execute the build command to compile your project, then start the server to test its functionality. Use curl to send a POST request to a sample endpoint to verify the application is running correctly.

```bash
npm run build

npm start
```

```bash
curl -X POST "http://127.0.0.1:3400/helloFlow" \
  -H "Content-Type: application/json" \
  -d '{"data": {"name": "Genkit"}}'
```

--------------------------------

### Install Genkit Astra DB Plugin

Source: https://genkit.dev/docs/js/integrations/astra-db

Install the Astra DB plugin for Genkit using npm.

```bash
npm install genkitx-astra-db
```

--------------------------------

### Install Cloud SQL PostgreSQL Plugin

Source: https://genkit.dev/docs/js/integrations/cloud-sql-postgresql

Install the Genkit Cloud SQL for PostgreSQL plugin using npm.

```bash
npm i --save genkitx-cloud-sql-pg
```

--------------------------------

### Example Cache Usage Metrics Output

Source: https://genkit.dev/docs/js/integrations/anthropic

An example of the structure of the cache usage metrics that can be found in the response metadata.

```typescript
{
  "inputTokens": 3,
  "outputTokens": 217,
  "custom": {
    "cache_creation_input_tokens": 3639,
    "cache_read_input_tokens": 0,
    "ephemeral_5m_input_tokens": 3639,
    "ephemeral_1h_input_tokens": 0
  }
}
```

--------------------------------

### Create a New Genkit Plugin Package

Source: https://genkit.dev/docs/js/plugin-authoring/overview

Provides the command-line steps to set up a new NPM package for a Genkit plugin, including creating the directory, initializing npm, and installing necessary dependencies like Genkit and TypeScript.

```bash
mkdir genkitx-my-plugin

cd genkitx-my-plugin

npm init -y

npm install genkit

npm install --save-dev typescript

npx tsc --init
```

--------------------------------

### Install PDF processing dependencies

Source: https://genkit.dev/docs/js/rag

Install the necessary libraries for PDF parsing, text chunking, and local vector storage.

```bash
npm install llm-chunk pdf-parse @genkit-ai/dev-local-vectorstore

npm install --save-dev @types/pdf-parse
```

--------------------------------

### Install LanceDB Genkit Plugin

Source: https://genkit.dev/docs/js/integrations/lancedb

Install the LanceDB plugin for Genkit using npm. This command is required before configuring the plugin.

```bash
npm install genkitx-lancedb
```

--------------------------------

### Initialize Genkit with a Plugin

Source: https://genkit.dev/docs/js/plugin-authoring/overview

Demonstrates how to import and configure Genkit with an existing plugin, such as the Vertex AI plugin, by passing it in the `plugins` array during initialization.

```typescript
import { genkit } from 'genkit';
import { vertexAI } from '@genkit-ai/vertexai';

const ai = genkit({
  plugins: [vertexAI({ projectId: 'my-project' })],
});
```

--------------------------------

### Start Genkit Developer UI

Source: https://genkit.dev/docs/js/app-frameworks/nuxt

Launch the Genkit Developer UI from your project root to test flows and inspect traces. The --watch flag ensures changes to your flow definitions are automatically reloaded.

```bash
npx genkit start -- tsx --watch server/utils/bargainChefFlow.ts
```

--------------------------------

### Install Genkit Package

Source: https://genkit.dev/docs/js/app-frameworks/sveltekit

Install the Genkit package for use in your SvelteKit project. This command is used when integrating with a standalone backend.

```bash
npm install genkit
```
Tokens:
Raw