### Clone Quick Start Example

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/quick-start.mdx

Fork the quick-start example locally to follow along with the guide. This command downloads the example project files.

```shell
npx git-ripper https://github.com/inngest/agent-kit/tree/main/examples/quick-start
```

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

### Run Multi-Agent Example

Source: https://github.com/inngest/agent-kit/blob/main/examples/mem0-memory/README.md

Execute the multi-agent example using npm. This command starts an agent setup that utilizes multiple agents for complex tasks.

```bash
npm run start-multi
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/examples/agentkit-starter/README.md

Clone the AgentKit repository, navigate to the starter example, and install project dependencies.

```bash
git clone https://github.com/inngest/agent-kit.git
cd agent-kit/examples/agentkit-starter

npm install

cp .env.example .env
```

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

### Navigate to Example Directory

Source: https://github.com/inngest/agent-kit/blob/main/examples/support-agent-human-in-the-loop/README.md

Change directory to the support agent example.

```bash
cd examples/support-agent-human-in-the-loop
```

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

### Navigate to Example Directory

Source: https://github.com/inngest/agent-kit/blob/main/examples/realtime-ui-nextjs/README.md

Change directory to the realtime-ui-nextjs example folder.

```bash
cd examples/realtime-ui-nextjs
```

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

### Start AgentKit Server

Source: https://github.com/inngest/agent-kit/blob/main/examples/swebench/README.md

Starts the AgentKit application server. This command should be run after dependencies are installed and API keys are set.

```shell
pnpm start
```

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

### Setup AgentKit Network and Agent

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/browserbase.mdx

Define an AgentKit agent and create a network to manage it. This example sets up a Reddit search agent.

```typescript
import {
  anthropic,
  createAgent,
  createNetwork,
} from "@inngest/agent-kit";

const searchAgent = createAgent({
  name: "reddit_searcher",
  description: "An agent that searches Reddit for relevant information",
  system:
  "You are a helpful assistant that searches Reddit for relevant information.",
});

// Create the network
const redditSearchNetwork = createNetwork({
  name: "reddit_search_network",
  description: "A network that searches Reddit using Browserbase",
  agents: [searchAgent],
  maxIter: 2,
  defaultModel: anthropic({
  model: "claude-3-5-sonnet-latest",
  max_tokens: 4096,
});

```

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

### Install Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/docs/README.md

Navigate to the docs directory and install project dependencies using npm.

```sh
cd docs
npm install
```

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

### Run the Agent Locally

Source: https://github.com/inngest/agent-kit/blob/main/examples/daytona-coding-agent/README.md

Execute this command to start the coding agent in your local environment after installing dependencies.

```bash
npm run start
```

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

### Initialize SWE-bench Examples

Source: https://github.com/inngest/agent-kit/blob/main/examples/swebench/README.md

Initializes the SWE-bench examples by running the 'make init' command. This typically downloads necessary data and prepares the environment for running benchmarks.

```shell
make init
```

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

### Code-based Router Example in Agent Kit

Source: https://github.com/inngest/agent-kit/blob/main/packages/agent-kit/README.md

Demonstrates setting up agents, tools, and a network with a code-based router. This setup orchestrates agents based on a plan generated by a code assistant agent, utilizing shared state for execution flow.

```typescript
import {
  z,
  anthropic,
  createAgent,
  createNetwork,
  createTool,
} from "@inngest/agent-kit";

import { readFileSync } from "fs";
import { join } from "path";

// create a shared tool
const saveSuggestions = createTool({
  name: "save_suggestions",
  description: "Save the suggestions made by other agents into the state",
  parameters: z.object({
    suggestions: z.array(z.string()),
  }),
  handler: async (input, { network }) => {
    const suggestions = network?.state.kv.get("suggestions") || [];
    network?.state.kv.set("suggestions", [
      ...suggestions,
      ...input.suggestions,
    ]);
    return "Suggestions saved!";
  },
});

// create agents with access to the state via the `saveSuggestions` tool
const documentationAgent = createAgent({
  name: "documentation_agent",
  system: "You are an expert at generating documentation for code",
  tools: [saveSuggestions],
});

const analysisAgent = createAgent({
  name: "analysis_agent",
  system: "You are an expert at analyzing code and suggesting improvements",
  tools: [saveSuggestions],
});

const summarizationAgent = createAgent({
  name: "summarization_agent",
  system: ({ network }) => {
    const suggestions = network?.state.kv.get("suggestions") || [];
    return `Save a summary of the following suggestions:
    ${suggestions.join("\n")}`;
  },
  tools: [
    createTool({
      name: "save_summary",
      description:
        "Save a summary of the suggestions made by other agents into the state",
      parameters: z.object({
        summary: z.string(),
      }),
      handler: async (input, { network }) => {
        network?.state.kv.set("summary", input.summary);
        return "Saved!";
      },
    }),
  ],
});

// Create the code assistant agent which generates a plan
const codeAssistantAgent = createAgent({
  name: "code_assistant_agent",
  system: ({ network }) => {
    const agents = Array.from(network?.agents.values() || [])
      .filter(
        (agent) =>
          !["code_assistant_agent", "summarization_agent"].includes(agent.name)
      )
      .map((agent) => `${agent.name} (${agent.system})`);
    return `From a given user request, ONLY perform the following tool calls:
- read the file content
- generate a plan of agents to run from the following list: ${agents.join(", ")}

Answer with "done" when you are finished.`;
  },
  tools: [
    createTool({
      name: "read_file",
      description: "Read a file from the current directory",
      parameters: z.object({
        filename: z.string(),
      }),
      handler: async (input, { network }) => {
        const filePath = join(process.cwd(), `files/${input.filename}`);
        const code = readFileSync(filePath, "utf-8");
        network?.state.kv.set("code", code);
        return "File read!";
      },
    }),
    createTool({
      name: "generate_plan",
      description: "Generate a plan of agents to run",
      parameters: z.object({
        plan: z.array(z.string()),
      }),
      handler: async (input, { network }) => {
        network?.state.kv.set("plan", input.plan);
        return "Plan generated!";
      },
    }),
  ],
});

const network = createNetwork({
  name: "code-assistant-v2",
  agents: [
    codeAssistantAgent,
    documentationAgent,
    analysisAgent,
    summarizationAgent,
  ],
  // our routing function relies on the shared state to orchestrate agents
  //   first, the codeAssistantAgent is called and then, its plan gets
  //   executed step by step until a summary gets written in the state.
  router: ({ network }) => {
    if (!network?.state.kv.has("code") || !network?.state.kv.has("plan")) {
      return codeAssistantAgent;
    } else {
      const plan = (network?.state.kv.get("plan") || []) as string[];
      const nextAgent = plan.pop();
      if (nextAgent) {
        network?.state.kv.set("plan", plan);
        return network?.agents.get(nextAgent);
      } else if (!network?.state.kv.has("summary")) {
        return summarizationAgent;
      } else {
        return undefined;
      }
    }
  },
  defaultModel: anthropic({
    model: "claude-3-5-sonnet-latest",
    defaultParameters: {
      max_tokens: 4096,
    },
  }),
});

```

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

### Install Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/README.md

Install all project dependencies using pnpm.

```bash
pnpm i
```

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

### Install @inngest/use-agent

Source: https://github.com/inngest/agent-kit/blob/main/packages/use-agent/README.md

Install the package using npm, pnpm, or yarn.

```bash
npm install @inngest/use-agent
# or
pnpm add @inngest/use-agent
# or
yarn add @inngest/use-agent
```

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

### Start Development Server

Source: https://github.com/inngest/agent-kit/blob/main/docs/README.md

Run this command to start a local development server for the documentation. It will be available at http://localhost:4321.

```sh
npm run dev
```

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

### Clone Repository and Navigate

Source: https://github.com/inngest/agent-kit/blob/main/examples/mcp-neon-agent/README.md

Navigate to the agent's example directory after cloning the repository.

```bash
cd examples/mcp-neon-agent
```

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

### Start Production Server

Source: https://github.com/inngest/agent-kit/blob/main/examples/deep-research/README.md

Launch the application in production mode after building.

```bash
pnpm run start
```

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

### Install AgentKit and Inngest with npm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/installation.mdx

Use this command to install both packages using npm.

```shell
npm install @inngest/agent-kit inngest
```

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

### Start Development Server

Source: https://github.com/inngest/agent-kit/blob/main/examples/deep-research/README.md

Run the Next.js development server to view the application locally.

```bash
pnpm run dev
```

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

### Download Example Code File

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/ai-workflows.mdx

Downloads the example.ts file from the specified GitHub URL using wget. This file is used by the Code Assistant example.

```bash
wget https://raw.githubusercontent.com/inngest/agent-kit/main/examples/code-assistant-rag/files/example.ts
```

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

### Install AgentKit and Daytona (npm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/daytona.mdx

Install the necessary packages for AgentKit and Daytona using npm.

```shell
npm install @inngest/agent-kit inngest @daytonaio/sdk
```

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

### Install Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/examples/e2b-coding-agent/README.md

Install project dependencies using npm. Ensure Node.js v16 or higher is installed.

```bash
npm install
```

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

### Install AgentKit and Inngest with yarn

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/installation.mdx

Use this command to install both packages using yarn.

```shell
yarn add @inngest/agent-kit inngest
```

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

### Install Peer Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/packages/use-agent/README.md

Ensure you have the required peer dependencies installed.

```bash
npm install react @inngest/realtime uuid
```

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

### Install Dependencies with yarn

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/ai-agents.mdx

Install the required agent-kit and zod packages using yarn.

```bash
yarn add @inngest/agent-kit zod
```

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

### Environment Setup Commit

Source: https://github.com/inngest/agent-kit/blob/main/examples/swebench/README.md

This snippet shows a commit hash used for environment setup.

```json
{
  "environment_setup_commit": "6072e0982c3c0236f532ddfa48fbf461180d834e"
}

```

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

### Install Dependencies with npm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/ai-agents.mdx

Install the required agent-kit, inngest, and zod packages using npm.

```bash
npm install @inngest/agent-kit inngest zod
```

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

### Install AgentKit and Daytona (yarn)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/daytona.mdx

Install the necessary packages for AgentKit and Daytona using yarn.

```shell
yarn add @inngest/agent-kit inngest @daytonaio/sdk
```

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

### Install AgentKit and Inngest with pnpm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/installation.mdx

Use this command to install both packages using pnpm.

```shell
pnpm install @inngest/agent-kit inngest
```

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

### Install AgentKit and E2B (npm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Install the necessary packages for AgentKit and E2B using npm.

```shell
npm install @inngest/agent-kit inngest @e2b/code-interpreter
```

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

### Install AgentKit and Daytona (pnpm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/daytona.mdx

Install the necessary packages for AgentKit and Daytona using pnpm.

```shell
pnpm install @inngest/agent-kit inngest @daytonaio/sdk
```

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

### Install AgentKit and Smithery SDK

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/smithery.mdx

Install the necessary packages for AgentKit and the Smithery SDK using npm, pnpm, or yarn.

```shell
npm install @inngest/agent-kit inngest @smithery/sdk
```

```shell
pnpm install @inngest/agent-kit inngest @smithery/sdk
```

```shell
yarn add @inngest/agent-kit inngest @smithery/sdk
```

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

### Install Dependencies with pnpm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/ai-agents.mdx

Install the required agent-kit, inngest, and zod packages using pnpm.

```bash
pnpm install @inngest/agent-kit inngest zod
```

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

### Install AgentKit and E2B (yarn)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Install the necessary packages for AgentKit and E2B using yarn.

```shell
yarn add @inngest/agent-kit inngest @e2b/code-interpreter
```

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

### Clone Repository and Navigate

Source: https://github.com/inngest/agent-kit/blob/main/examples/voice-assistant/README.md

Clone the AgentKit repository and navigate to the voice-assistant example directory.

```sh
git clone https://github.com/inngest/agent-kit.git
cd agent-kit/examples/voice-assistant
```

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

### Start Qdrant Vector Database

Source: https://github.com/inngest/agent-kit/blob/main/examples/mem0-memory/README.md

Start the Qdrant local vector database using Docker Compose. This is required for storing and retrieving memories.

```bash
docker-compose up
```

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

### Install Inngest SDK with npm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/concepts/deployment.mdx

Use npm to install the Inngest SDK. This is the first step in integrating Inngest with your AgentKit network.

```shell
npm install inngest
```

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

### Run AgentKit Server

Source: https://github.com/inngest/agent-kit/blob/main/examples/gemini-quick-start/README.md

Start the AgentKit server using this command.

```shell
npm start
```

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

### Environment Variables Setup

Source: https://github.com/inngest/agent-kit/blob/main/examples/reddit-search-browserbase-tools/README.md

Configure your Browserbase and Anthropic API keys in a .env file.

```env
BROWSERBASE_API_KEY=your_browserbase_api_key
BROWSERBASE_PROJECT_ID=your_browserbase_project_id
ANTHROPIC_API_KEY=your_anthropic_api_key
```

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

### Environment Variables Setup

Source: https://github.com/inngest/agent-kit/blob/main/examples/use-agent/README.md

Configure environment variables for API keys and Inngest credentials in a .env.local file.

```env
# LLM Provider API Keys (at least one required)
OPENAI_API_KEY=your-openai-api-key
ANTHROPIC_API_KEY=your-anthropic-api-key

# Inngest (optional for local development)
INNGEST_EVENT_KEY=your-inngest-event-key
INNGEST_SIGNING_KEY=your-inngest-signing-key
```

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

### Install AgentKit and Inngest

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/quick-start.mdx

Install AgentKit and its required peer dependency, Inngest, using npm, pnpm, or yarn.

```shell
npm install @inngest/agent-kit inngest
```

```shell
pnpm install @inngest/agent-kit inngest
```

```shell
yarn add @inngest/agent-kit inngest
```

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

### Download Example Code File

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/agentic-workflows.mdx

Downloads the example TypeScript file 'example.ts' into a 'files' directory. This file is used by the code assistant for analysis.

```bash
mkdir files
cd files
wget https://raw.githubusercontent.com/inngest/agent-kit/main/examples/code-assistant-agentic/files/example.ts
cd -
```

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

### MCP Agent Example with Neon Database

Source: https://github.com/inngest/agent-kit/blob/main/packages/agent-kit/README.md

Example demonstrating how to build an AgentKit Agent using MCP with the Neon database MCP server. This requires setting up a Smithery URL with your profile and API key.

```tsx
import {
  anthropic,
  createAgent,
  createNetwork,
  createTool,
} from "@inngest/agent-kit";
import { createServer } from "@inngest/agent-kit/server";
import { createSmitheryUrl } from "@smithery/sdk/config.js";
import { z } from "zod";

const neonServerUrl = createSmitheryUrl(
  "https://server.smithery.ai/neon/mcp?profile=bored-catshark-da8hf&api_key=123123123"
);

const neonAgent = createAgent({
  name: "neon-agent",
  system: `You are a helpful assistant that help manage a Neon account.
  IMPORTANT: Call the 'done' tool when the question is answered.
  `,
  tools: [
    createTool({
      name: "done",
      description: "Call this tool when you are finished with the task.",
      parameters: z.object({
        answer: z.string().describe("Answer to the user's question."),
      }),
      handler: async ({ answer }, { network }) => {
        network?.state.kv.set("answer", answer);
      },
    }),
  ],
  mcpServers: [
    {
      name: "neon",
      transport: {
        type: "streamable-http",
        url: neonServerUrl.toString(),
      },
    },
  ],
});

const neonAgentNetwork = createNetwork({
  name: "neon-agent",
  agents: [neonAgent],
  defaultModel: anthropic({
    model: "claude-3-5-sonnet-20240620",
    defaultParameters: {
      max_tokens: 1000,
    },
  }),
  router: ({ network }) => {
    if (!network?.state.kv.get("answer")) {
      return neonAgent;
    }
    return;
  },
});

// Create and start the server
const server = createServer({
  networks: [neonAgentNetwork],
});

server.listen(3010, () =>
  console.log("Support Agent demo server is running on port 3010")
);

```

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

### Example Handler with Network State Update

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/reference/create-tool.mdx

This example demonstrates a tool handler that interacts with the agent's network state. It updates `network.state.fileWritten` after successfully writing a file.

```typescript
import { createTool } from '@inngest/agent-kit';

const tool = createTool({
  name: 'write-file',
  description: 'Write a file to disk with the given contents',
  parameters: {
    type: 'object',
    properties: {
      path: { type: 'string' },
      contents: { type: 'string' },
    },
  },
  handler: async ({ path, contents }, { agent, network }) => {
    await fs.writeFile(path, contents);
    network.state.fileWritten = true;
    return { success: true };
  },
});
```

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

### Static System Prompt Example

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/concepts/agents.mdx

Defines an agent with a simple, static system prompt for a copy editor.

```ts
const codeWriterAgent = createAgent({
  name: 'Copy editor',
  system:
    `You are an expert copy editor. Given a draft article, you provide ` +
    `actionable improvements for spelling, grammar, punctuation, and formatting.`,
  model: openai('gpt-3.5-turbo'),
});
```

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

### Install AgentKit and Inngest

Source: https://github.com/inngest/agent-kit/blob/main/README.md

Install AgentKit and Inngest dependencies. Starting with AgentKit v0.9.0, `inngest` must be installed separately.

```bash
npm i @inngest/agent-kit inngest
```

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

### Set Up Environment Variables

Source: https://github.com/inngest/agent-kit/blob/main/examples/voice-assistant/README.md

Copy the example environment file and populate it with your API keys for services like Anthropic, Google Maps, Notion, and Exa.

```sh
cp .env.example .env
```

```env
ANTHROPIC_API_KEY=sk-ant-xxx
GOOGLE_MAPS_API_KEY=AIzaSy...
NOTION_API_KEY=secret_...
EXA_API_KEY=...
```

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

### Build and Preview Documentation

Source: https://github.com/inngest/agent-kit/blob/main/docs/README.md

Build the documentation for production and then preview the build locally.

```sh
npm run build
npm run preview  # preview the build locally
```

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

### Implement Conversation History Adapter

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/concepts/history.mdx

Define a history adapter by implementing the `createThread`, `get`, `appendUserMessage`, and `appendResults` methods. This example demonstrates saving to a PostgreSQL database.

```typescript
import {
  createNetwork,
  createAgent,
  createState,
  openai,
} from "@inngest/agent-kit";
import { db } from "./db"; // Your database client

// Define your history adapter with all four methods
const conversationHistoryAdapter: HistoryConfig<any> = {
  // 1. Create new conversation threads (or ensure they exist)
  createThread: async ({ state, input }) => {
    // If a threadId already exists, upsert to ensure it's in the database
    if (state.threadId) {
      await db.thread.upsert({
        where: { id: state.threadId },
        update: { updatedAt: new Date() },
        create: {
          id: state.threadId,
          userId: state.data.userId,
          title: input.slice(0, 50),
          createdAt: new Date(),
        },
      });
      return { threadId: state.threadId };
    }

    // Otherwise, create a new thread
    const thread = await db.thread.create({
      data: {
        userId: state.data.userId,
        title: input.slice(0, 50), // First 50 chars as title
        createdAt: new Date(),
      },
    });
    return { threadId: thread.id };
  },

  // 2. Load conversation history (including user messages)
  get: async ({ threadId }) => {
    if (!threadId) return [];

    const messages = await db.message.findMany({
      where: { threadId },
      orderBy: { createdAt: "asc" },
    });

    // Transform ALL messages (user + agent) to AgentResult format
    // This preserves the complete conversation order
    return messages.map((msg) => {
      if (msg.role === "user") {
        // Convert user messages to AgentResult with agentName: "user"
        return new AgentResult(
          "user",
          [
            {
              type: "text" as const,
              role: "user" as const,
              content: msg.content,
              stop_reason: "stop",
            },
          ],
          [],
          new Date(msg.createdAt)
        );
      } else {
        // Return agent results
        return new AgentResult(
          msg.agentName,
          [
            {
              type: "text" as const,
              role: "assistant" as const,
              content: msg.content,
            },
          ],
          [],
          new Date(msg.createdAt)
        );
      }
    });
  },

  // 3. Save user message immediately (before agents run)
  appendUserMessage: async ({ threadId, userMessage }) => {
    if (!threadId) return;

    await db.message.create({
      data: {
        messageId: userMessage.id, // Use canonical client-generated ID
        threadId,
        role: "user",
        content: userMessage.content,
        createdAt: userMessage.timestamp,
      },
    });
  },

  // 4. Save agent results after the run
  appendResults: async ({ threadId, newResults }) => {
    if (!threadId) return;

    // Save only agent responses (user message already saved)
    for (const result of newResults) {
      const content = result.output
        .filter((msg) => msg.type === "text")
        .map((msg) => msg.content)
        .join("\n");

      await db.message.create({
        data: {
          messageId: result.id || crypto.randomUUID(), // Use result.id if available
          threadId,
          role: "assistant",
          agentName: result.agentName,
          content,
          checksum: result.checksum, // For idempotency
          createdAt: result.createdAt,
        },
      });
    }
  },
};

```

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

### Daytona Coding Agent Setup

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/daytona.mdx

Initializes an AgentKit coding agent and imports necessary tools for interacting with the Daytona sandbox. This snippet is a partial example showing the agent creation and import of Daytona SDK components.

```typescript
import {
  createAgent,
  anthropic,
  createTool,
} from "@inngest/agent-kit";
import { z } from "zod";
import { CodeRunParams, DaytonaError } from "@daytonaio/sdk";
import { getSandbox, logDebug } from "./utils.js";

const codingAgent = createAgent({
  name: "Coding Agent",
  description:
    "An autonomous coding agent for building software in a Daytona sandbox",

```

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

### Hybrid Code and Agent Routing

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/concepts/routers.mdx

Combine predictable code-based routing for initial steps with agent-based routing for subsequent decisions. This example uses code to always start with the classifier, then delegates to the default routing agent.

```typescript
import { createNetwork, getDefaultRoutingAgent } from "@inngest/agent-kit";

// classifier and writer Agents definition...

const network = createNetwork({
  agents: [classifier, writer],
  router: ({ callCount }) => {
    // Always start with the classifier
    if (callCount === 0) {
      return classifier;
    }
    // Then let the routing agent take over
    return getDefaultRoutingAgent();
  },
});
```

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

### Install Dependencies with pnpm

Source: https://github.com/inngest/agent-kit/blob/main/examples/voice-assistant/README.md

Install project dependencies using pnpm. Ensure pnpm is installed globally.

```sh
pnpm install
```

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

### Create a New Project

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/smithery.mdx

Initialize a new project using npm, pnpm, or yarn if you don't have an existing one.

```shell
mkdir my-agent-kit-project && npm init
```

```shell
mkdir my-agent-kit-project && pnpm init
```

```shell
mkdir my-agent-kit-project && yarn init
```

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

### Run Network with User Prompt (Example)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/agentic-workflows.mdx

Demonstrates how to initiate the agent network's execution with a user-provided prompt for code analysis. The network orchestrates agents and shares state.

```typescript
await network.run(
  `Analyze the files/example.ts file by suggesting improvements and documentation.`
);
```

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

### Set Up Environment Variables

Source: https://github.com/inngest/agent-kit/blob/main/examples/deep-research/README.md

Configure necessary API keys for OpenAI and Exa by creating a .env file.

```env
# Required: OpenAI API key for AI model inference
OPENAI_API_KEY=your_openai_api_key_here

# Required: Exa API key for web search in deep research
EXA_API_KEY=your_exa_api_key_here
```

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

### Create New Project (npm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Initialize a new project with npm if you don't have an existing one.

```shell
mkdir my-agent-kit-project && npm init
```

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

### Configure Environment Variables

Source: https://github.com/inngest/agent-kit/blob/main/examples/e2b-coding-agent/README.md

Set up your E2B and Anthropic API keys by copying the example environment file and filling in your credentials.

```env
E2B_API_KEY=your_e2b_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key
```

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

### Initialize Database Schema

Source: https://github.com/inngest/agent-kit/blob/main/examples/agentkit-starter/README.md

Run the provided script to set up the necessary tables in your PostgreSQL database.

```bash
npm run setup-db
```

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

### Create New Project (pnpm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Initialize a new project with pnpm if you don't have an existing one.

```shell
mkdir my-agent-kit-project && pnpm init
```

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

### Start the Inngest CLI Dev Server

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/getting-started/quick-start.mdx

Launch the Inngest development server and connect it to your AgentKit server by providing the URL.

```shell
npx inngest-cli@latest dev -u http://localhost:3000/api/inngest
```

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

### Initialize Project with npm

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/guided-tour/ai-workflows.mdx

Initializes a new Node.js project using npm. This is a prerequisite for installing project dependencies.

```bash
npm init
```

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

### Start Development Servers

Source: https://github.com/inngest/agent-kit/blob/main/examples/agentkit-starter/README.md

Run the Inngest CLI in development mode and the Next.js application concurrently in separate terminals.

```bash
# Terminal 1: Start Inngest dev server
npx inngest-cli@latest dev
```

```bash
# Terminal 2: Start Next.js app
pnpm run dev
```

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

### Create New Project (yarn)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Initialize a new project with yarn if you don't have an existing one.

```shell
mkdir my-agent-kit-project && yarn init
```

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

### Install AgentKit and E2B (pnpm)

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/integrations/e2b.mdx

Install the necessary packages for AgentKit and E2B using pnpm.

```shell
pnpm install @inngest/agent-kit inngest @e2b/code-interpreter
```

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

### Function Router Example

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/reference/network-router.mdx

An example demonstrating how to define a function router for the `defaultRouter` option in `createNetwork`.

```APIDOC
## Function Router

A function router is provided to the `defaultRouter` option in `createNetwork`.

### Example

```ts
const network = createNetwork({
  agents: [classifier, writer],
  router: ({ lastResult, callCount, network, stack, input }) => {
    // First call: use the classifier
    if (callCount === 0) {
      return classifier;
    }

    // Get the last message from the output
    const lastMessage = lastResult?.output[lastResult?.output.length - 1];
    const content =
      lastMessage?.type === "text" ? (lastMessage?.content as string) : "";

    // Second call: if it's a question, use the writer
    if (callCount === 1 && content.includes("question")) {
      return writer;
    }

    // Otherwise, we're done!
    return undefined;
  },
});
```

### Parameters

- **input** (string) - The original input provided to the network.
- **network** (Network) - The network instance, including its state and history.
  See [`Network.State`](/reference/state) for more details.
- **stack** (Agent[]) - The list of future agents to be called. (_internal read-only value_)
- **callCount** (number) - The number of agent calls that have been made.
- **lastResult** (InferenceResult) - The result from the previously called agent.
  See [`InferenceResult`](/reference/state#inferenceresult) for more details.

### Return Values

| Return Type    | Description                                        |
| -------------- | -------------------------------------------------- |
| `Agent`        | Single agent to execute next                       |
| `Agent[]`      | Multiple agents to execute in sequence             |
| `RoutingAgent` | Delegate routing decision to another routing agent |
| `undefined`    | Stop network execution                             |
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/examples/deep-research/README.md

Clone the AgentKit repository and install project dependencies using pnpm.

```bash
git clone <repository-url>
cd agentkit-chat
pnpm install
```

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

### Install @inngest/use-agent and Peer Dependencies

Source: https://github.com/inngest/agent-kit/blob/main/docs/src/content/docs/reference/react-hooks/overview.mdx

Install the @inngest/use-agent package along with its peer dependencies: react, @inngest/realtime, and uuid.

```bash
npm install @inngest/use-agent
# Peer dependencies
npm install react @inngest/realtime uuid
```

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

### Example Agent Prompt

Source: https://github.com/inngest/agent-kit/blob/main/examples/daytona-coding-agent/README.md

This TypeScript code snippet shows how to initiate the agent with a prompt to create a React application. The agent will then scaffold, build, and potentially run the application.

```typescript
const result = await network.run(
  `Create a minimal React app called "Notes" that lets users add, view, and delete notes. Each note should have a title and content. Use Create React App or Vite for setup. Include a simple UI with a form to add notes and a list to display them.`
);
```