### AIGNE Basic Quick Start Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/getting-started/quick-start.md

Demonstrates the fundamental steps of using the AIGNE framework. It initializes the AIGNE core, defines an AI agent with specific instructions, and invokes the agent to get a response. It also includes an example of handling streaming responses from the agent.

```typescript
import { AIAgent, AIGNE, isAgentResponseDelta } from "@aigne/core";
import { OpenAIChatModel } from "@aigne/openai";

const aigne = new AIGNE({
  model: new OpenAIChatModel({
    apiKey: process.env.OPENAI_API_KEY,
    model: "gpt-4o-mini",
  }),
});

const agent = AIAgent.from({
  instructions: "You are a helpful assistant",
  inputKey: "message",
});

const result = await aigne.invoke(agent, { message: "What is AIGNE?" });
console.log(result);
// Output: { message: "AIGNE is a platform for building AI agents." }

const stream = await aigne.invoke(
  agent,
  { message: "What is AIGNE?" },
  { streaming: true },
);

let response = "";
for await (const chunk of stream) {
  console.log(chunk);
  if (isAgentResponseDelta(chunk) && chunk.delta.text?.message)
    response += chunk.delta.text.message;
}
console.log(response);
// Output:  "AIGNE is a platform for building AI agents."
```

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

### Run the Example (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/memory-did-spaces/README.md

This command starts the AIGNE Framework example after dependencies are installed and environment variables are configured.

```bash
pnpm start
```

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

### Install Dependencies and Setup Environment (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-reflection/README.md

Commands to navigate to the example directory, install dependencies using pnpm, and set up the OpenAI API key in a local environment file.

```bash
cd aigne-framework/examples/workflow-reflection

pnpm install

OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Install Dependencies and Setup Environment

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-puppeteer/README.md

Navigate to the example directory, install dependencies using pnpm, and set up the OpenAI API key in a local environment file.

```bash
cd aigne-framework/examples/mcp-puppeteer

pnpm install

# Setup environment variables
# Create or edit .env.local file
# OPENAI_API_KEY="YOUR_OPENAI_API_KEY"
```

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

### Install AIGNE Core and OpenAI Package

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/getting-started/quick-start.md

Installs the core AIGNE framework package and the OpenAI integration package using npm. This is the first step to enable AI agent capabilities.

```bash
npm install @aigne/core @aigne/openai
```

```bash
yarn add @aigne/core @aigne/openai
```

```bash
pnpm add @aigne/core @aigne/openai
```

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

### Quick Start: Run AIGNE Workflow Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-code-execution/README.md

Provides commands to quickly run the AIGNE Framework's code-execution workflow example without local installation. It covers setting the OpenAI API key and running in one-shot, interactive chat, or with pipeline input.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY # Set your OpenAI API key

# Run in one-shot mode (default)
npx -y @aigne/example-workflow-code-execution

# Run in interactive chat mode
npx -y @aigne/example-workflow-code-execution --chat

# Use pipeline input
echo 'Calculate 15!' | npx -y @aigne/example-workflow-code-execution
```

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

### Install Dependencies and Run Example (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-group-chat/README.md

Navigates to the group chat example directory, installs dependencies using pnpm, and runs the example in one-shot or interactive mode.

```bash
cd aigne-framework/examples/workflow-group-chat

pnpm install

pnpm start # Run in one-shot mode (default)

# Run in interactive chat mode
pnpm start -- --chat

# Use pipeline input
echo "Write a short story about space exploration" | pnpm start
```

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

### Install Dependencies and Setup Environment (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-orchestrator/README.md

Navigates to the workflow orchestrator example directory, installs dependencies using pnpm, and sets up the OpenAI API key in a local environment file.

```bash
cd aigne-framework/examples/workflow-orchestrator

pnpm install

OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Run the Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-puppeteer/README.md

Start the AIGNE Framework Puppeteer MCP Server example using the pnpm start command.

```bash
pnpm start
```

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

### Install Dependencies for MCP Blocklet Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-blocklet/README.md

Navigate to the MCP Blocklet example directory and install its dependencies using pnpm.

```bash
cd aigne-framework/examples/mcp-blocklet

pnpm install
```

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

### Install Dependencies (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/memory-did-spaces/README.md

This command navigates into the example directory and installs the necessary project dependencies using pnpm.

```bash
cd aigne-framework/examples/memory-did-spaces

pnpm install
```

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

### Quick Start Chatbot Execution (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/chat-bot/README.md

Demonstrates how to quickly run the AIGNE chatbot example using npx without local installation. It shows one-shot mode, interactive chat mode, and using pipeline input.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

# Run in one-shot mode (default)
npx -y @aigne/example-chat-bot

# Run in interactive chat mode
npx -y @aigne/example-chat-bot --chat

# Use pipeline input
echo "Tell me about AIGNE Framework" | npx -y @aigne/example-chat-bot
```

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

### Install Dependencies (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-did-spaces/README.md

This command navigates into the MCP DID Spaces example directory within the cloned AIGNE Framework repository and installs the necessary dependencies using pnpm.

```bash
cd aigne-framework/examples/mcp-did-spaces

pnpm install
```

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

### Install Dependencies for MCP GitHub Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-github/README.md

Installs project dependencies using pnpm after navigating to the example directory. Assumes the repository has been cloned.

```bash
cd aigne-framework/examples/mcp-github

pnpm install
```

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

### Install Dependencies for Memory Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/memory/README.md

Installs the necessary Node.js dependencies for the AIGNE Framework memory example using pnpm. This command should be run after navigating into the example's directory.

```bash
cd aigne-framework/examples/memory

pnpm install
```

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

### Install Dependencies with pnpm

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-code-execution/README.md

Installs project dependencies using pnpm after navigating to the example's directory within the cloned repository.

```bash
cd aigne-framework/examples/workflow-code-execution

pnpm install
```

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

### Run AIGNE GitHub MCP Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-github/README.md

Quick start command to run the AIGNE example for GitHub MCP Server integration. Requires setting environment variables for OpenAI API key and GitHub token.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY
export GITHUB_TOKEN=YOUR_GITHUB_TOKEN

npx -y @aigne/example-mcp-github
```

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

### Bash: Install Dependencies with pnpm

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-handoff/README.md

Commands to navigate into the example directory and install project dependencies using pnpm. This is required after cloning the repository.

```bash
cd aigne-framework/examples/workflow-handoff

pnpm install
```

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

### Install Dependencies for AIGNE Workflow (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-sequential/README.md

Navigates to the workflow-sequential example directory and installs project dependencies using pnpm.

```bash
cd aigne-framework/examples/workflow-sequential

pnpm install
```

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

### Install Dependencies for AIGNE Workflow Concurrency

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-concurrency/README.md

Navigate to the example directory and install project dependencies using pnpm.

```bash
cd aigne-framework/examples/workflow-concurrency

pnpm install
```

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

### Install AIGNE CLI (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/chat-bot/README.md

Provides the command to globally install the AIGNE CLI tool using npm.

```bash
npm install -g @aigne/cli
```

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

### Install Dependencies (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-sqlite/README.md

Installs project dependencies using pnpm after navigating to the example directory.

```bash
cd aigne-framework/examples/mcp-sqlite
pnpm install
```

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

### Run AIGNE Memory Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/memory/README.md

Starts the AIGNE Framework memory example application using pnpm. This command executes the main script for the chatbot with memory capabilities.

```bash
pnpm start
```

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

### Setup Environment Variables (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-did-spaces/README.md

This section provides instructions on setting up environment variables for the MCP DID Spaces example. It includes copying a configuration file, and then editing it to add API keys and server details for OpenAI and DID Spaces.

```bash
# Copy the example configuration
cp CONFIG.md .env.local

# Edit the file and add your API keys
OPENAI_API_KEY=your_openai_api_key_here
DID_SPACES_URL=https://your-did-spaces-url.com/app/mcp
DID_SPACES_AUTHORIZATION=Bearer your-did-spaces-token
```

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

### Start AIGNE Project

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs-examples/test-aigne/README.md

Starts the AIGNE project using the installed CLI. This command typically launches the chat agent and any associated services defined in the project configuration.

```bash
aigne run
```

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

### Setup Environment Variables for AIGNE

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-code-execution/README.md

Instructions for setting up the OpenAI API key in a local environment file for the AIGNE Framework example.

```bash
OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Create a Simple AI Agent

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/getting-started/quick-start.md

Defines a basic AI agent using the `AIAgent.from()` method. It specifies the agent's core instructions and the key for its input.

```typescript
const agent = AIAgent.from({
  instructions: "You are a helpful assistant",
  inputKey: "message",
});
```

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

### Install Dependencies with pnpm (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-router/README.md

Navigates to the workflow-router example directory and installs project dependencies using pnpm.

```bash
cd aigne-framework/examples/workflow-router

pnpm install
```

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

### Complete Example: Build and Invoke First AIGNE Agent

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/build-your-first-agent.md

A comprehensive TypeScript example demonstrating the complete lifecycle of creating a basic AI Agent with AIGNE. It includes importing modules, setting up the AIGNE instance with an OpenAI model, configuring the agent, and invoking it to get a response.

```typescript
import { AIAgent, AIGNE } from "@aigne/core";
import { OpenAIChatModel } from "@aigne/openai";

const aigne = new AIGNE({
  model: new OpenAIChatModel({
    apiKey: process.env.OPENAI_API_KEY,
    model: "gpt-4o-mini",
  }),
});

const agent = AIAgent.from({
  instructions: "You are a helpful assistant for Crypto market cap",
  inputKey: "message",
});

const result = await aigne.invoke(agent, { message: "What is crypto?" });
console.log(result);
// Output: { message: "Cryptocurrency, often referred to as crypto, is a type of digital or virtual currency that uses cryptography for security" }
```

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

### Configure Environment Variables

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/complete-cli-workflow.md

Copies the example environment file and instructs users to add their API keys for different AI model providers like OpenAI, Anthropic, or Google.

```bash
cd my-ai-assistant
cp .env.local.example .env.local

# OpenAI API Key
OPENAI_API_KEY=your_openai_api_key_here

# Or other model provider keys
# ANTHROPIC_API_KEY=your_anthropic_api_key_here
# GOOGLE_API_KEY=your_google_api_key_here
```

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

### Run Memory Chatbot with Quick Start Commands

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/memory/README.md

Demonstrates how to quickly run the memory-enabled chatbot using npx commands. It shows how to set the OpenAI API key and interact with the chatbot by providing input or starting an interactive chat session.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

# Run the chatbot with memory
npx -y @aigne/example-memory --input 'I like blue color'
npx -y @aigne/example-memory --input 'What is my favorite color?'

# Run in interactive chat mode
npx -y @aigne/example-memory --chat
```

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

### Adding Example to Index

Source: https://github.com/aigne-io/aigne-framework/blob/main/CONTRIBUTING-EXAMPLES.md

Illustrates how to add a new example to the main examples index in the `examples/README.md` file, maintaining alphabetical order.

```Markdown
- [@aigne/example-your-name: Brief description](./your-example-name/README.md)
```

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

### Setup Environment Variables (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-sqlite/README.md

Sets up the OpenAI API key in a local environment file for the AIGNE Framework example.

```bash
OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Run MCP Blocklet Example with pnpm

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-blocklet/README.md

Start the MCP Blocklet example using pnpm. Supports default one-shot mode or specifying a Blocklet app URL.

```bash
pnpm start

pnpm start https://your-blocklet-app-url
```

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

### Run AIGNE Chatbot with AIGNE CLI (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/chat-bot/README.md

Demonstrates how to run the AIGNE chatbot example using the locally installed AIGNE CLI. Includes commands for one-shot mode, interactive chat, and pipeline input.

```bash
aigne run # Run in one-shot mode (default)

# Run in interactive chat modeaigne run --chat

# Use pipeline input
echo "Tell me about AIGNE Framework" | aigne run
```

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

### Setup Environment Variables (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-router/README.md

Sets up the OpenAI API key in the .env.local file for the AIGNE Framework example.

```bash
OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Bash: Quick Start Workflow Handoff

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-handoff/README.md

Provides commands to quickly run the AIGNE Framework workflow handoff demo without local installation. It includes setting the OpenAI API key and executing the demo in one-shot or chat modes, with pipeline input.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY # Set your OpenAI API key

# Run in one-shot mode (default)
npx -y @aigne/example-workflow-handoff

# Run in interactive chat mode
npx -y @aigne/example-workflow-handoff --chat

# Use pipeline input
echo "transfer to agent b" | npx -y @aigne/example-workflow-handoff
```

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

### Install AIGNE CLI

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/complete-cli-workflow.md

Installs the AIGNE CLI globally using npm and verifies the installation by checking the CLI version.

```bash
npm install -g @aigne/cli

aigne --version
```

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

### Invoke Agent with Guide Rails in TypeScript

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/concepts/agent.md

Demonstrates invoking an Agent that has guide rails applied. It shows an example where a user's query about Bitcoin prices is intercepted and handled according to the guide rail's restrictions.

```TypeScript
const result = await aigne.invoke(agent, {
  message: "What will be the price of Bitcoin next month?",
});
console.log(result);
// Output:
// {
//   "message": "I cannot provide cryptocurrency price predictions as they are speculative and potentially misleading."
// }
```

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

### Start MCP Server

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-server/README.md

Starts the AIGNE MCP server on a specified port. Requires setting the OPENAI_API_KEY environment variable. The server provides an endpoint for MCP-compatible clients.

```bash
OPENAI_API_KEY="" # Set your OpenAI API key here

# Start the MCP server
npx -y @aigne/example-mcp-server serve-mcp --port 3456

# Output
# Observability OpenTelemetry SDK Started, You can run `npx aigne observe` to start the observability server.
# MCP server is running on http://localhost:3456/mcp
```

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

### Complete Client-Side Example (TypeScript)

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/serve-agent-as-api.md

This comprehensive example illustrates the complete client-side implementation for invoking Agent services. It includes importing the client, configuring the connection URL, retrieving an agent, and sending a message.

```typescript
import { AIGNEHTTPClient } from "@aigne/transport/http-client/index.js";

const client = new AIGNEHTTPClient({
  url: `http://localhost:${port}/api/chat`,
});

const chatbot = await client.getAgent({ name: "chatbot" });
const result = await chatbot.invoke({
  message: "What is the crypto price of ABT/USD on coinbase?",
});
console.log(result);
// Output: { message: "The current price of ABT/USD on Coinbase is $0.9684." }
```

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

### Run Group Chat Example with Options (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-group-chat/README.md

Demonstrates running the AIGNE Framework group chat example with various command-line options such as setting the chat mode, log level, and providing input.

```bash
# Run in chat mode (interactive)
pnpm start -- --chat

# Set logging level
pnpm start -- --log-level DEBUG

# Use pipeline input
echo "Write a short story about space exploration" | pnpm start
```

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

### Start AIGNE Monitoring Service

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/complete-cli-workflow.md

Starts the AIGNE monitoring service on the default port 7890. It can also be configured to use a custom port or allow public access via a specified host.

```bash
aigne observeaigne observe --port 8080aigne observe --host 0.0.0.0
```

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

### Clone Repository and Navigate (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/chat-bot/README.md

Shows how to clone the AIGNE framework repository and navigate to the chatbot example directory.

```bash
git clone https://github.com/AIGNE-io/aigne-framework

cd aigne-framework/examples/chat-bot
```

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

### Full Example: Add Cryptocurrency Skills to Agent

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/add-skills-to-agent.md

A complete example demonstrating the setup of an AIGNE instance, creation of an MCPAgent for cryptocurrency data, integration of this skill into an AIAgent, and invoking the Agent to retrieve market data.

```TypeScript
import { AIAgent, AIGNE, MCPAgent } from "@aigne/core";
import { OpenAIChatModel } from "@aigne/openai";

const aigne = new AIGNE({
  model: new OpenAIChatModel(),
});

const ccxt = await MCPAgent.from({
  command: "npx",
  args: ["-y", "@mcpfun/mcp-server-ccxt"],
});

const agent = AIAgent.from({
  instructions: "You are a helpful assistant for Crypto market cap",
  skills: [ccxt],
  inputKey: "message",
});

const result = await aigne.invoke(agent, {
  message: "What is the crypto price of ABT/USD on coinbase?",
});
console.log(result);
// Output: { message:"The current price of ABT/USD on Coinbase is $0.9684." }
```

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

### Quick Start: Run MCP Blocklet Demo

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-blocklet/README.md

Execute the MCP Blocklet Demo using npx. Supports one-shot mode (default), interactive chat mode, and pipeline input.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY
export BLOCKLET_APP_URL="https://xxx.xxxx.xxx"

# Run in one-shot mode (default)
npx -y @aigne/example-mcp-blocklet

# Run in interactive chat mode
npx -y @aigne/example-mcp-blocklet --chat

# Use pipeline input
echo "What are the features of this blocklet app?" | npx -y @aigne/example-mcp-blocklet
```

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

### Bash: Setup Environment Variables

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-handoff/README.md

Instructions for setting up the OpenAI API key in a local environment file. This is crucial for the framework to authenticate with OpenAI services.

```bash
OPENAI_API_KEY="" # Set your OpenAI API key here
```

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

### Run AIGNE Chatbot Example

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/README.md

Demonstrates how to run the AIGNE chatbot example using npx. It shows both one-shot mode and interactive chat mode, requiring an OpenAI API key to be set as an environment variable.

```bash
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

# Run in one-shot mode
npx -y @aigne/example-chat-bot

# Or add "--chat" parameter to enter interactive chat mode
npx -y @aigne/example-chat-bot --chat
```

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

### Invoke AI Agent for a Response

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/getting-started/quick-start.md

Executes the AI agent with a specific input and retrieves a response from the AIGNE framework. The result is logged to the console.

```typescript
const result = await aigne.invoke(agent, { message: "What is AIGNE?" });
console.log(result);
// Output: { message: "AIGNE is a platform for building AI agents." }
```

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

### AIGNE CLI Basic Commands

Source: https://github.com/aigne-io/aigne-framework/blob/main/packages/cli/README.md

Provides a quick reference for essential @aigne/cli commands, including displaying help, creating new projects, running agents, executing tests, serving MCP services, and starting the observability server.

```bash
# Display help informationaigne --help

# Create a new projectaigne create [path]

# Run an agentaigne run --path xxx

# Run testsaigne test --path xxx

# Start MCP serveraigne serve-mcp --path xxx

# Start observability serveraigne observe [option]
```

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

### Observe Agents

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-server/README.md

Starts the AIGNE observability server on a specified port, allowing monitoring of agent interactions and performance metrics via a web dashboard.

```bash
npx aigne observe --port 7890
```

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

### Use Pipeline Input with aigne-framework

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-github/README.md

This example shows how to pipe input into the aigne-framework using `echo` and `pnpm start`. It searches for repositories related to 'modelcontextprotocol'.

```bash
echo "Search for repositories related to 'modelcontextprotocol'" | pnpm start
```

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

### Create New AIGNE Project

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/how-to-guides/complete-cli-workflow.md

Creates a new AIGNE project either interactively or by specifying a project name. The CLI then sets up the project directory and initial configuration files.

```bash
# Interactive project creationaigne create

# Or specify project nameaigne create my-ai-assistant
```

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

### Stream AI Agent Responses

Source: https://github.com/aigne-io/aigne-framework/blob/main/docs/getting-started/quick-start.md

Invokes the AI agent in streaming mode to receive responses incrementally. It processes each chunk and concatenates the text message parts.

```typescript
const stream = await aigne.invoke(
  agent,
  { message: "What is AIGNE?" },
  { streaming: true },
);

let response = "";
for await (const chunk of stream) {
  console.log(chunk);
  if (isAgentResponseDelta(chunk) && chunk.delta.text?.message)
    response += chunk.delta.text.message;
}
console.log(response);
// Output:  "AIGNE is a platform for building AI agents."
```

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

### Setup Environment Variables for AIGNE

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/mcp-github/README.md

Configures environment variables for API keys (OpenAI and GitHub) in a local .env file for the AIGNE framework example.

```bash
OPENAI_API_KEY=""
GITHUB_TOKEN=""
```

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

### AIGNE Workflow Sequential Demo Run Options (Bash)

Source: https://github.com/aigne-io/aigne-framework/blob/main/examples/workflow-sequential/README.md

Examples of running the AIGNE Framework's sequential workflow with different command-line options, including chat mode, log level, and piped input.

```bash
# Run in chat mode (interactive)
pnpm start -- --chat

# Set logging level
pnpm start -- --log-level DEBUG

# Use pipeline input
echo "Create marketing content for our new AI-powered fitness app" | pnpm start
```

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

### Run AIGNE Project

Source: https://github.com/aigne-io/aigne-framework/blob/main/packages/core/test-agents/README.md

Starts the AIGNE project. This command typically launches the chat agent and makes its functionalities available.

```bash
aigne run
```